A non-stationary reservoir hydrological drought identification method, device, medium and product

Abstract

The application discloses a non-stationary reservoir hydrological drought identification method and device, medium and product, and relates to the technical field of hydrology and drought identification. The method comprises the following steps: performing detrended processing on historical original storage data of a target reservoir in a historical time period; calculating the average value of the detrended daily storage corresponding to the Lth day and the S days before the Lth day in the daily storage sequence; dividing the average detrended daily storage of the same date into a group, and calculating a plurality of preliminary threshold values of drought quantiles of each group; calculating the average value of each preliminary threshold value of drought quantiles in the Dth day and the M days before and after the Dth day in a year, as the drought threshold value of the Dth day, to obtain a drought threshold value group of each day; and determining the drought event of the target reservoir in the target year according to the comparison value of the detrended daily storage of each day of the target year and the corresponding same-day drought threshold value of each day. The application improves the accuracy of non-stationary reservoir hydrological drought identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hydrology and drought identification, and particularly relates to a non-stationary reservoir hydrological drought identification method, device, medium and product. BACKGROUND

[0002] Hydrological drought has a great impact on water resources management, agricultural production and ecological environment. According to the water storage state, hydrological drought can be divided into reservoir hydrological drought, river hydrological drought and groundwater hydrological drought. Reservoir is an important source of regional water supply and is greatly affected by regional water allocation, and has strong non-stationarity.

[0003] Traditional drought identification methods are mostly based on fixed thresholds or static statistical analysis, which cannot effectively deal with non-stationary and seasonal hydrological data, resulting in insufficient identification accuracy, and the quantitative evaluation of drought characteristics is often lack of accuracy and dynamics. SUMMARY

[0004] The purpose of the present application is to provide a non-stationary reservoir hydrological drought identification method, device, medium and product, which can improve the accuracy of non-stationary reservoir hydrological drought identification.

[0005] To achieve the above purpose, the present application provides the following solutions:

[0006] The first aspect of the present application provides a non-stationary reservoir hydrological drought identification method, which comprises: performing detrended processing on historical original storage data of a target reservoir in a historical time period to obtain a detrended daily storage sequence; calculating the average value of the detrended daily storage corresponding to the Lth day and the S days before the Lth day in the detrended daily storage sequence as the average detrended daily storage of the Lth day, to obtain a full sequence of daily storage; wherein the full sequence of daily storage comprises the average detrended daily storage of each day in the detrended daily storage sequence; dividing the average detrended daily storage of the same date in the full sequence of daily storage into a group, and calculating a plurality of drought quantile preliminary thresholds of each group to obtain a group of drought quantile preliminary thresholds of each day in the full sequence of daily storage; the group of drought quantile preliminary thresholds comprises a plurality of drought quantile preliminary thresholds; calculating the average value of each drought quantile preliminary threshold in the group of drought quantile preliminary thresholds at the Dth day in the year and the M days before and after the Dth day as the drought threshold of the Dth day, to obtain a group of drought thresholds of each day; the group of drought thresholds comprises a plurality of drought thresholds; determining the drought event of the target reservoir in the target year according to the comparison value of the detrended daily storage of each day in the target year and the corresponding same date of each drought threshold.

[0007] According to the embodiment of the present application, the above-mentioned de-trending the historical raw storage data of the target reservoir in a historical time period to obtain a de-trending daily storage sequence comprises: applying a Gaussian filtering method to de-trend the historical raw storage data of the target reservoir in a historical time period to obtain a de-trending daily storage sequence.

[0008] According to the embodiment of the present application, the above-mentioned calculating the average value of the de-trending daily storage corresponding to the Lth day and the S days before the Lth day in the de-trending daily storage sequence as the average de-trending daily storage of the Lth day to obtain the full sequence of daily storage comprises: when the interval days between the Lth day and the starting date in the de-trending daily storage sequence are greater than or equal to S, calculating the average value of the de-trending daily storage corresponding to the Lth day and the S days before the Lth day in the de-trending daily storage sequence to obtain the daily storage of the Lth day; when the interval days between the Lth day and the starting date in the de-trending daily storage sequence are less than S, calculating the average value of the de-trending daily storage corresponding to the Lth day and the interval days between the Lth day and the starting date in the de-trending daily storage sequence to obtain the daily storage of the Lth day; and obtaining the full sequence of daily storage according to the daily storage of the Lth day.

[0009] According to the embodiment of the present application, the above-mentioned dividing the average de-trending daily storage of the same date in the full sequence of daily storage into a group and calculating a plurality of preliminary threshold values of drought quantiles of each group to obtain a group of preliminary threshold values of drought quantiles of each day in the full sequence of daily storage comprises: dividing the average de-trending daily storage of the same date in each year in the full sequence of daily storage into a group; calculating the first, second and third percentiles for a plurality of average de-trending daily storages of each group, and taking the first percentile as a preliminary threshold value of heavy drought, the second percentile as a preliminary threshold value of moderate drought, and the third percentile as a preliminary threshold value of light drought; and the group of preliminary threshold values of drought quantiles of each day in the full sequence of daily storage comprises the preliminary threshold value of heavy drought, the preliminary threshold value of moderate drought and the preliminary threshold value of light drought.

[0010] According to the embodiment of the present application, the above method further comprises: extracting a characteristic value of the drought event, and generating a corresponding statistical table according to the characteristic value and the drought event.

[0011] According to the embodiment of the present application, the above method further comprises: extracting a characteristic value of the drought event, and generating a corresponding statistical table according to the characteristic value and the drought event.

[0012] According to the embodiment of the present application, the above method further comprises: extracting a characteristic value of the drought event, and generating a corresponding statistical table according to the characteristic value and the drought event.

[0013] The second aspect of the present application provides a computer device, comprising: a memory, a processor to store a computer program on the memory and run the computer program on the processor, and the processor executes the computer program to realize the non-stationary reservoir hydrological drought identification method.

[0014] The third aspect of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the non-stationary reservoir hydrological drought identification method.

[0015] The fourth aspect of the present application provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to realize the non-stationary reservoir hydrological drought identification method.

[0016] According to the embodiment of the present application, the above method further comprises: extracting a characteristic value of the drought event, and generating a corresponding statistical table according to the characteristic value and the drought event.

[0017] The application provides a non-stationary reservoir hydrological drought identification method. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0019] Figure 1 A flowchart of the non-stationary reservoir hydrological drought identification method provided by the present application is shown.

[0020] Figure 2 A curve diagram of reservoir original storage data, Gaussian filter trend items and detrended storage provided by the present application is shown.

[0021] Figure 3 A threshold curve diagram for daily judgment of drought after smoothing provided by the present application is shown.

[0022] Figure 4 A comparison diagram of reservoir hydrological drought events identified by the non-stationary reservoir hydrological drought identification method and reservoir hydrological drought events identified by directly using original storage sequence is shown.

[0023] Figure 5 An internal structure diagram of the computer device is shown. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort fall within the scope of protection of the present application.

[0025] At present, hydrological drought identification methods are mainly divided into two categories: stationary methods and non-stationary methods. The stationary method assumes that the statistical characteristics (such as mean, variance, etc.) of the hydrological sequence remain unchanged over time, while the non-stationary method considers the characteristics of the hydrological sequence changing over time, which can better reflect the influence of climate change and human activities on the hydrological system. Common non-stationary methods include:

[0026] 1. Time-varying threshold method: This method dynamically adjusts the threshold based on the long-term trend or seasonal changes of the hydrological series, avoiding the limitations of fixed thresholds. Common methods include sliding window quantiles, adaptive quantiles, and seasonal factor correction.

[0027] 2. Time series decomposition method: The hydrological series is decomposed and analyzed separately.

[0028] 3. Machine learning-based methods: Using machine learning models (such as support vector machines, random forests, etc.) to capture the nonlinear characteristics of hydrological sequences.

[0029] Although non-stationary methods have shown good adaptability in hydrological drought identification, they still have some limitations:

[0030] 1. Insufficient decomposition granularity: Existing methods are mostly based on monthly data for time series decomposition, but hydrological drought events may be triggered by short-cycle fluctuations. Coarse-grained decomposition will mask key details, leading to errors in the description of the temporal characteristics of drought.

[0031] 2. Rigid threshold setting: Existing non-stationary methods rely heavily on the fixed quantile values ​​of the original sequence when setting thresholds, without considering the coupling effect between the trend term and the residual term, resulting in insufficient matching between the threshold and the actual drought conditions.

[0032] 3. Overfitting risk in trend fitting: For non-stationary sequences, detrending methods are often used to obtain stationary sequences before identification. Traditional trend extraction methods may lead to large fluctuations in the fitted curve due to insufficient control of model complexity. Abnormal fluctuations in the trend line can distort the determination of the intensity and duration of drought events, thus leading to systematic bias in drought attribution analysis.

[0033] The purpose of this invention is to provide a method, device, medium, and product for identifying non-stationary reservoir hydrological drought. By detrending the daily storage data of the reservoir, a daily variable drought threshold is constructed to identify and assess reservoir hydrological drought events. Combining techniques such as time series decomposition, moving average, and quantile threshold analysis, it can accurately and effectively identify non-stationary reservoir hydrological drought and calculate its characteristic values, which is of great significance for the accurate identification of reservoir hydrological drought.

[0034] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0035] Example 1

[0036] like Figure 1 As shown, the non-stationary reservoir hydrological drought identification method in this embodiment includes:

[0037] Step S1: Detrend the historical raw storage data of the target reservoir in a historical time period to obtain a detrended daily storage sequence.

[0038] Specifically, the Gaussian filtering method is applied to detrend the historical raw storage data of the target reservoir in a historical time period to obtain a detrended daily storage sequence.

[0039] In actual application, the raw storage data is first detrended by the Gaussian filtering method, the standard deviation parameter (σ) of the filter is set to 731, and the boundary processing parameter (mode) is set to "reflect mode" to obtain a long-term trend item. Then, the smoothed trend item is subtracted from the original daily storage data to generate a detrended daily storage sequence. In this step, the raw storage data is historical data that has been measured, and can cover any time period, but it is recommended to use continuous data to ensure the accuracy of the smoothed trend.

[0040] Gaussian filtering is a filtering method commonly used for signal smoothing and trend extraction, and can be used to extract the long-term change trend of the storage sequence in hydrological time series analysis, thereby realizing detrend processing and adapting to the needs of non-stationary drought identification. In Gaussian filtering, each original daily storage data has a corresponding smoothed trend item. Gaussian filtering is a filtering operation for each data point to generate a smoothed long-term trend, and after the trend is subtracted, the detrended data can be obtained. Each smoothed trend item is based on the entire time series, so its shape is the same (i.e., it is a trend processed by Gaussian filtering), but its value at different time points is different. The trend item generated by Gaussian filtering is dynamically changing, and is not a constant. Therefore, the value of the smoothed trend item at each time point is different.

[0041] The Gaussian kernel function expression is as follows:

[0042] (1)

[0043] wherein, represents the weight at a distance of t time from the center; is the standard deviation of the Gaussian kernel, which controls the width of the smoothing window; t is the time offset relative to the center point. In the present application, 731 is taken.

[0044] Given the original storage , the smoothed trend part is calculated by the following formula:

[0045] (2)

[0046] wherein 2n+1 represents the width of the sliding window, and in actual application, the value is determined according to the actual situation Selecting a proper window size.

[0047] After the smoothing process is completed, the trend term can be removed from the original sequence to obtain a detrended storage sequence

[0048] (3)

[0049] wherein t is the index value of the daily data, and represents the tth day in the data, Table

[0050] detrended data of the tth day, represents the original number of the tth day

[0051] According to represents the trend part of the tth day obtained by Gaussian filtering and smoothing.

[0052] The sequence calculated does not contain a long-term trend term, and can be used in the subsequent drought identification process.

[0053] Step S2: calculating the average value of the detrended daily storage corresponding to the Lth day and the S days before the Lth day in the detrended daily storage sequence as the average detrended daily storage of the Lth day, to obtain a daily storage sequence; wherein the daily storage sequence includes the average detrended daily storage of each day in the detrended daily storage sequence.

[0054] S2 includes:

[0055] Step S21: when the interval days between the Lth day and the starting date in the detrended daily storage sequence are greater than or equal to S, calculating the average value of the detrended daily storage corresponding to the Lth day and the S days before the Lth day in the detrended daily storage sequence to obtain the daily storage of the Lth day.

[0056] Step S22: when the interval days between the Lth day and the starting date in the detrended daily storage sequence are less than S, calculating the average value of the detrended daily storage corresponding to the Lth day and the interval days between the Lth day and the starting date in the detrended daily storage sequence to obtain the daily storage of the Lth day.

[0057] Step S23: obtaining the daily storage sequence based on the daily storage of the Lth day.

[0058] In actual application, the value of t is the Lth day in the daily sequence, and based on the detrended storage data, the average value is calculated by taking the storage of the previous 30 days (including the current day) of each day t as the basis, and a new sequence is obtained by rolling average.​ For the detrended daily storage data, t is the index value of the daily data, indicating the tth day in the data. The specific calculation steps are as follows:

[0059] When t>31 (i.e. the 31st day and after), for each date t, the rolling average is obtained by calculating the average of the data of the day and the previous 30 days:

[0060] (4)

[0061] Where t represents the current tth day, and t-30 represents 30 days before the current date (including the current date). The rolling average window includes the daily storage data of the current date and the previous 30 days.

[0062] When t<31 (i.e. from the 1st day to the 30th day), since there are not enough 30 days of data, the rolling average calculation will use all the data from the 1st day to the current date t:

[0063] (5)

[0064] In the case of t<31, the rolling average is the average of all data from the 1st day to the current date t. In this way, when the first 31 days (including the current day) are averaged, the situation of not having enough data for 31 days at the beginning is considered, avoiding missing values at the beginning of the sequence and ensuring that there is a result for the entire sequence. Therefore, in the present application, t represents L, and the value of S can be selected as 30.

[0065] Step S3: dividing the average detrended daily storage of the same date in the full sequence of daily storage into a group, and calculating a plurality of preliminary threshold values of drought quantiles for each group to obtain a preliminary threshold value group of drought quantiles for each day in the full sequence of daily storage; the preliminary threshold value group of drought quantiles includes a plurality of preliminary threshold values of drought quantiles.

[0066] S3 specifically includes:

[0067] Step S31: dividing the average detrended daily storage of the same date in each year in the full sequence of daily storage into a group.

[0068] Step S32: for a plurality of average detrended daily storages of each group, calculating the first, second and third percentiles, and taking the first percentile as the preliminary threshold value of heavy drought, the second percentile as the preliminary threshold value of moderate drought, and the third percentile as the preliminary threshold value of light drought; the preliminary threshold value group of drought quantiles for each day in the full sequence of daily storage includes the preliminary threshold value of heavy drought, the preliminary threshold value of moderate drought, and the preliminary threshold value of light drought.

[0069] Specifically, the smoothed sequence is ​Grouping by "calendar day", i.e. collecting data of the same date in all years, and calculating the 6th, 16th, 30th percentiles of the daily storage for each date.

[0070] Due to the small amount of data of leap years, errors are prone to occur, therefore, in the present application, when grouping by "calendar day" and calculating the 6th, 16th, 30th percentiles of each group of data after grouping, the data of leap years are not considered, and the data of leap years are set as: The day sequence of the calendar day (Date), each group of data is:

[0071] (6)

[0072] wherein, The date corresponding to the tth day, for each calendar day D, the quantile preliminary threshold of the residual sequence thereof is calculated, and is defined as follows:

[0073] The preliminary threshold of severe drought (6th percentile, corresponding to the standardized index threshold value -1.5):

[0074] (7)

[0075] The preliminary threshold of moderate drought (16th percentile, corresponding to the standardized index threshold value -1):

[0076] (8)

[0077] The preliminary threshold of light drought (30th percentile, corresponding to the standardized index threshold value -0.5):

[0078] (9)

[0079] Finally, a group of daily drought quantile preliminary thresholds covering 365 days of the year is formed

[0080]

[0081] The 6th percentile represents the preliminary threshold of severe drought (approximately corresponding to the standardized index threshold value -1.5), the 16th percentile represents the preliminary threshold of moderate drought (approximately corresponding to the standardized index threshold value -1), and the 30th percentile serves as the preliminary threshold of light drought (approximately corresponding to the standardized index threshold value -0.5).

[0082] Step S4: calculating the average value of each drought quantile preliminary threshold in the group of drought quantile preliminary thresholds in the Dth day of the day sequence number in a year and the M days before and after the Dth day, as the drought threshold of the Dth day, to obtain a group of drought thresholds of each day; the drought threshold group includes a plurality of drought thresholds.

[0083] S4 includes:

[0084] Step S41: Calculate the average of the light drought preliminary threshold value in the Dth day, the M days before and after the Dth day, and obtain the light drought threshold value.

[0085] Step S42: Calculate the average of the moderate drought preliminary threshold value in the Dth day, the M days before and after the Dth day, and obtain the moderate drought threshold value.

[0086] Step S43: Calculate the average of the severe drought preliminary threshold value in the Dth day, the M days before and after the Dth day, and obtain the severe drought threshold value, and take the light drought threshold value, the moderate drought threshold value and the severe drought threshold value as the drought threshold value group of date T.

[0087] In actual application, the 31-day moving average algorithm is applied to the daily drought threshold value of different levels, and the average of the 31-day storage capacity before and after the day is taken to form the smooth daily drought threshold value from January 1 to December 31.

[0088] (10)

[0089] wherein, represents the specific percentile corresponding to the preliminary threshold value of drought of the level

[0090] D represents the day sequence in a year. When the 31-day moving average algorithm is applied, M can be 15, and the 31-day is composed of the Dth day, the M days before and after the Dth day.

[0091] In the formula, the annual cycle filling method is used when the index exceeds the range:

[0092] For dates before D=1 (such as D=-1), the end data (such as D=364) is used for filling. For dates after D=365, the beginning data (such as D=1, 2, …) is used for filling.

[0093] For the smooth threshold value of February 29, interpolation is used before and after the day, and the calculation formula is as follows:

[0094] (11)

[0095] wherein, is the threshold value corresponding to the level of February 29, and are the threshold values corresponding to February 28 and March 1, respectively. This data is suitable for interpolation filling of February 29 in leap years to ensure that the threshold value sequence of 366 days in a year is continuous and complete.

[0096] The final obtained ​respectively, constitute a complete judgment basis. When used to calculate The data used to calculate The data used to calculate ; D represents the day sequence in a year.

[0097] Step S5: Determine the drought event of the target reservoir in the target year according to the comparison value of the detrended daily storage of each day of the target year and the corresponding drought threshold value of the same date.

[0098] S5 includes:

[0099] Step S51: Obtain the drought identification result of each day of the target reservoir in the target year according to the comparison value of the detrended daily storage of each day of the target year and the corresponding drought threshold value of the same date; the drought identification result includes no drought and drought; the drought level includes light drought, moderate drought and severe drought.

[0100] Step S52: When the drought identification result of each day of the continuous time period in the target year is drought, a drought event occurs in the continuous time period.

[0101] Step S53: When the time interval between adjacent two drought events does not exceed the preset interval days, the adjacent two drought events are combined into one drought event.

[0102] In practical application, based on the detrended daily storage sequence , compare with the smoothed daily quantile threshold value , wherein, is the day sequence in a year on the t-th day, determine whether a day is a drought day, if the storage of a day is less than the threshold value of the date, mark it as a drought day:

[0103] If , the day is a drought day (12)

[0104] , wherein, corresponding to severe drought, moderate drought and light drought, t is the index value of the daily data, indicating the t-th day in the data, is the corresponding calendar day sequence of the t-th day, indicating the actual position of the day in a year, in a common year, , in a leap year, .

[0105] ​The drought event identification will be based on the heavy drought, moderate drought and light drought threshold values corresponding to the 6th, 16th and 30th percentile d values, respectively, to extract drought events of different levels to comprehensively reflect the spatial and temporal variation characteristics of drought levels.

[0106] If all the days are drought days for 20 days or more, the period is identified as a drought event. If the interval between two drought periods does not exceed 5 days (including 5 days), the two drought events are combined into one drought.

[0107] In addition, the non-stationary reservoir hydrological drought identification method further comprises:

[0108] The characteristic values of the drought events are extracted, and a corresponding statistical table is generated according to the characteristic values and the drought events.

[0109] In actual application, for each drought event (indexed as ), the following characteristic values are calculated:

[0110] Drought start time : The occurrence time of the drought event, i.e. the first day time in the drought sequence that meets the drought condition:

[0111] (13)

[0112] wherein, is the set of indexes of all drought days in the drought event .

[0113] Drought end time : The last time point at which the reservoir storage recovers to above the threshold value and no longer meets the drought condition.

[0114] (14)

[0115] Drought center time : The midpoint of the duration of the drought event, calculated by adding the start time to half of the drought duration.

[0116] (15)

[0117] Drought duration : The number of days of drought, the length of time from the start to the end of the drought.

[0118] (16)

[0119] Drought intensity : The maximum deficit of the reservoir storage relative to the threshold value during the drought event. The maximum deficit refers to the maximum value among the differences between the daily reservoir storage and the corresponding drought threshold value (such as the moderate drought threshold value) during a certain drought event.

[0120] (17)

[0121] Intensity of drought : Accumulated deficit amount of storage relative to threshold during drought event. Intensity of drought is the deficit amount of the most severe day of drought, emphasizing "maximum deficit" and measuring the degree of the most severe drought. Intensity of drought is the sum of deficit amount of each day during drought, which can also be understood as "total deficit amount", emphasizing "overall loss", and intensity is more affected by duration than intensity.

[0122] (18)

[0123] Frequency of drought : Relative duration of drought event in the year to which it belongs, and the calculation method is drought duration divided by the number of days in the year to which the center time of the drought event belongs. Let be the year to which the center time of the drought belongs, and the number of days in this year is (generally 365 or 366):

[0124] (19)

[0125] The annual attribution method adopted by the frequency of drought is to take the year to which the center time of the drought event belongs as the statistical attribution year, for example, assuming that there is a drought event:

[0126] Duration: December 20, 2022 to January 10, 2023, a total of 22 days. The center day of drought falls on December 30, 2022. Then the drought event belongs to 2022. The frequency of the event is calculated as: 22 / 365.

[0127] Finally, the detailed information table of the drought event is output, including: drought event year sequence number (according to the starting time to be attributed to the corresponding year), starting time, ending time, drought duration, drought intensity, drought intensity and frequency, which provides a scientific basis for subsequent water resources management and drought warning.

[0128] The application is a non-stationary reservoir hydrological drought identification method based on a detrended variable drought threshold, which comprises the following steps:

[0129] (1) Original data detrending: calculate the annual mean of daily storage data to generate a yearly sequence. The trend item is extracted by Gaussian filtering smoothing. Then, the trend item is deducted from the original data to generate a detrended storage sequence. In this embodiment, daily storage data of a reservoir in a certain place from 2001 to 2023 is selected for research. The specific steps are as follows:

[0130] Gaussian filtering smoothing: the original storage data First, the detrended processing is performed by using the Gaussian filter method, and the standard deviation parameter (σ) of the filter is set to 731; the boundary processing parameter (mode) is set to "reflect mode" (reflect), to obtain the long-term trend item T(t).

[0131] Generating the detrended storage sequence: subtracting the trend item from the original daily storage data to generate the detrended storage sequence .

[0132] (2) Calculating the daily different quantile storage of the sequence: in this embodiment, based on the generated post-detrended storage, a new sequence is formed after smoothing , the multi-year data is grouped and counted, and the different quantile values of the daily storage are calculated for each day. In this step, the 6th, 16th and 30th percentiles are selected to reflect different levels of storage. The specific steps are as follows:

[0133] Smoothed detrended storage: based on the detrended storage data, the average of the storage of the previous 30 days (including the current day) is taken day by day, and a new sequence is obtained by rolling average .

[0134] Grouping by year and date: grouping the data by year and date to obtain 365 groups of data excluding February 29 .

[0135] Calculating quantiles: for each group, calculate multiple statistical values for the day, including:

[0136] 6th percentile : The 6th percentile of the daily storage corresponds to the standardized index threshold value -1.5), which is used to identify severe drought events.

[0137] 16th percentile : The 16th percentile of the daily storage (corresponding to the standardized index threshold value -1), which is used to identify moderate drought events.

[0138] 30th percentile : The 30th percentile of the daily storage (corresponding to the standardized index threshold value -0.5), which is used to identify light drought events.

[0139] Selection of quantiles: these quantile storage values can be used to identify the threshold of drought events. In this process, since the value of February 29 is small and prone to error, the value of February 29 is not considered.

[0140] (3) 31-day moving average is performed on the different quantile storage to form the daily threshold for judging drought:

[0141] Based on the different quantile series calculated in the second step, further smoothing is performed using a 31-day moving average method. The specific steps are as follows:

[0142] Moving average calculation: Apply a 31-day moving average algorithm to take the average of the daily storage volume centered on the current day and the previous and next 31 days, smoothing the excessive fluctuations of the daily drought threshold, forming a smooth daily drought threshold from January 1 to December 31. Considering the case of leap years, the threshold value for February 29 is obtained by interpolating the different quantile values for February 28 and March 1.

[0143] (4) Based on the daily quantile threshold of the moving average, identify the reservoir hydrological drought and calculate the reservoir hydrological drought characteristic value:

[0144] The identification of reservoir hydrological drought is usually done by comparing with the drought threshold, using the smoothed storage sequence generated in the second step, and based on the run theory, the daily quantile threshold obtained in the third step is used as the basis for drought determination. The specific steps are as follows:

[0145] Identify drought days: If the storage volume of a certain day is less than the threshold value of that day, it is considered as a drought day.

[0146] Drought period start and end: from the day when the storage volume first falls below the threshold value , mark the beginning of the drought event. When the storage volume recovers to the threshold value and above, and the drought period exceeds a certain number of days (20 days), the drought is considered to be over.

[0147] Drought merging: If the interval between two drought periods does not exceed 5 days (including 5 days), the two drought events are merged into one drought.

[0148] Calculation of drought characteristic value: For each drought event, calculate the characteristic value to quantify the severity, duration, etc. of the drought. The specific calculation method is as follows:

[0149] Drought start time : The occurrence time of the drought event, i.e. the first day of the drought sequence that meets the drought conditions.

[0150] Drought end time : The last time point when the drought event storage volume recovers to the threshold value and no longer meets the drought conditions.

[0151] Drought center time : The midpoint of the duration of the drought event, calculated by adding half of the duration to the start time.

[0152] Drought duration : the number of days of drought, the length of time from the beginning to the end of drought.

[0153] drought intensity : the maximum deficit amount of storage relative to the threshold value during the drought event.

[0154] drought intensity : the cumulative deficit amount of storage relative to the threshold value during the drought event.

[0155] drought frequency : the relative duration of the drought event in the year to which it belongs, calculated by dividing the drought duration by the number of days in the year to which the center time of the drought event belongs. Let be the year to which the center time of the drought belongs, and the number of days in the year is (usually 365 or 366).

[0156] Result output: after the identification and feature calculation of the drought event are completed, all drought events and their feature values are summarized and output as a light drought table, a moderate drought table and a severe drought table.

[0157] Compared with the reservoir hydrological drought results of the reservoir storage before detrending, the original data can hardly identify drought events after 2015, while the detrended data can still identify drought events, as shown in Tables 1, 2 and 3. It proves that the method has applicability to non-stationary data.

[0158] Table 1: Light drought event of a reservoir in a certain year

[0159]

[0160] Table 2: Moderate drought event of a reservoir in a certain year

[0161]

[0162] Table 3: Severe drought event of a reservoir in a certain year

[0163]

[0164] The present application effectively solves the interference of trend interference and seasonal fluctuations on drought identification in the traditional method through Gaussian filter detrending processing and dynamic threshold construction. Specifically, first, the original reservoir daily storage data is subjected to Gaussian filter detrending processing, and the stationary storage sequence is obtained by deducting the long-term trend item Secondly, based on the rolling 30-day (including the current day) storage mean sequence A preliminary threshold system of daily quantiles was constructed (the 6th, 16th, and 30th percentiles correspond to severe / moderate / mild drought), and the threshold curve was smoothed by a moving 31-day average filled with annual cycle data; finally, based on the generated thresholds... Drought events were identified using runs theory, and their duration was calculated. intensity strength frequency Equivalent feature values. Compared with the traditional fixed threshold method, this invention innovatively couples trend stripping with dynamic thresholds, eliminates non-stationary trend interference through Gaussian filtering, and maintains the continuity of the threshold within the year by combining moving average, achieving dual correction of historical trends and seasonal fluctuations. This significantly improves the spatiotemporal accuracy of drought identification and provides a more scientific quantitative basis for reservoir drought risk assessment.

[0165] Figure 2 The original reservoir storage data, Gaussian filtered trend term, and detrended storage curve provided for this invention.

[0166] like Figure 2 As shown, the original storage data and trend term plot display the original storage data of the reservoir (gray solid line) and the trend term after Gaussian filtering (black dashed line). The detrended storage plot displays the storage of the reservoir after removing the trend term (black solid line). Observing the trend term dashed line, it can be seen that the storage of a certain reservoir showed an upward trend from 2001 to 2023. Comparing the storage curves before and after detrending, it can be seen that the detrending method using Gaussian filtering is effective and feasible.

[0167] Figure 3 The smoothed threshold curve for daily drought assessment is provided for the present invention.

[0168] like Figure 3 As shown, the square marker line represents the threshold for mild drought after a 31-day moving average, the triangle marker line represents the threshold for moderate drought after a 31-day moving average, and the cross marker line represents the threshold for severe drought after a 31-day moving average. The dynamic thresholds generated by quantiles and moving averages can adapt to seasonal changes in water storage, providing a scientific basis for drought level determination.

[0169] Figure 4 This is a comparison chart showing the occurrence times of reservoir hydrological drought events identified using the non-stationary reservoir hydrological drought identification method provided by this invention, and those identified directly using the original storage sequence.

[0170] like Figure 4As shown, the gray column chart in the figure is the reservoir hydrological drought event identified by the original storage data, and the black column chart is the reservoir hydrological drought event identified by the detrended storage, and the comparison result shows that the original storage data cannot identify drought events after 2015, and the drought events identified by the detrended data are more uniform in time distribution, proving that the application has good identification effect when processing data with obvious trend, and has good applicability to non-stationary data.

[0171] The application detrends the storage data by Gaussian filtering, extracts the long-term trend by Gaussian smoothing, and removes the trend term from the original data. After detrending, the detrended data is combined with moving average and quantile threshold analysis, which can effectively identify reservoir hydrological drought events and quantify the characteristic values of drought (such as drought duration, drought intensity and drought strength). Compared with traditional methods, the present method avoids the influence of long-term trend on quantile calculation by Gaussian filtering detrending, captures seasonal changes within a year by daily calculation of quantiles, makes the threshold more accurate, and avoids systematic bias in the drought identification process caused by trend interference. In addition, the application quantifies the characteristic values of drought duration, intensity and strength, etc., to provide reliable data support for quantitative analysis of drought events, thereby helping to optimize water resource management strategies.

[0172] Embodiment 2

[0173] A computer device, comprising: a memory, a processor to store a computer program on the memory and executable on the processor, and the processor executes the computer program to realize the non-stationary reservoir hydrological drought identification method in embodiment 1.

[0174] Embodiment 3

[0175] A computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to realize the non-stationary reservoir hydrological drought identification method in embodiment 1.

[0176] Embodiment 4

[0177] A computer program product comprising a computer program, the computer program being executed by a processor to realize the non-stationary reservoir hydrological drought identification method in embodiment 1.

[0178] Embodiment 5

[0179] A computer device, which can be a database, and its internal structure diagram can be as follows Figure 5As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is used to store the to-be-processed transaction. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with the terminal outside through the network connection. The computer program is executed by the processor to implement the non-stationary reservoir hydrological drought identification method in embodiment 1.

[0180] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0181] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0182] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0183] The principles and implementation modes of the present application are described by using specific examples in this paper, and the above-mentioned embodiments are only used to help understand the method and core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed. In conclusion, the content of the present application should not be understood as a limitation.

Claims

1. A non-stationary reservoir hydrological drought identification method, characterized in that, The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises:

2. The non-stationary reservoir hydrologic drought identification method of claim 1, wherein, The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: The method comprises: When the interval between the Lth day and the starting date in the detrended daily storage sequence is less than S, the average value of the detrended daily storage on the Lth day and the detrended daily storage corresponding to the interval between the Lth day and the starting date in the detrended daily storage sequence is calculated to obtain the daily storage on the Lth day; According to the daily storage on the Lth day, the full sequence of daily storages is obtained.

3. The non-stationary reservoir hydrologic drought identification method of claim 1, wherein, The comparison between the detrended daily storage on each day of the target year and the corresponding drought threshold value on the same date is used to determine the drought event of the target reservoir in the target year, including: According to the comparison between the detrended daily storage on each day of the target year and the corresponding drought threshold value on the same date, the drought identification result of the target reservoir on each day of the target year is obtained; the drought identification result includes no drought and drought; the drought level includes light drought, moderate drought, and severe drought; When the drought identification result of each day in a continuous time period of the target year is drought, a drought event occurs in the continuous time period; When the time interval between two adjacent drought events does not exceed the preset interval days, the two adjacent drought events are combined into one drought event.

4. The non-stationary reservoir hydrologic drought identification method of claim 3, wherein, The method further includes: The characteristic value of the drought event is extracted, and a corresponding statistical table is generated according to the characteristic value and the drought event.

5. A computer apparatus comprising: The memory and the processor are used to store a computer program on the memory and run the computer program on the processor, and the processor executes the computer program to implement the non-stationary reservoir hydrological drought identification method according to any one of claims 1-4.

6. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the non-stationary reservoir hydrological drought identification method according to any one of claims 1-4.

7. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the non-stationary reservoir hydrological drought identification method according to any one of claims 1-4. The computer program is executed by the processor to implement the non-stationary reservoir hydrological drought identification method according to any one of claims 1-4.

Images