Reservoir ecological scheduling method for simulating natural flood peak

Abstract

The invention discloses a reservoir ecological scheduling method for simulating a natural flood peak, and the method comprises the steps: (1) carrying out runoff detection through employing a multi-variation diagnosis method, so as to restore natural runoff; (2) solving an optimal flood characteristic value by adopting a multi-probability density method, determining a flood characteristic value threshold in combination with a guarantee rate, and enhancing the applicability of an ecological scheduling demand index; and (3) verifying and optimizing the ecological scheduling scheme of the reservoir in Dengping low water years, and increasing the scientificity of the ecological scheduling scheme. According to the method, the flood characteristic requirement of the natural river is considered, the reliability of the river ecological hydrological requirement is improved, and research results can provide a theoretical basis for reservoir ecological scheduling and provide necessary reference for river water resource management.

Description

Technical Field

[0001] The invention relates to the technical field of ecological dispatching of hydropower projects, and in particular to a reservoir ecological dispatching method simulating natural flood peaks. Background Art

[0002] Ecological regulation research is of great significance in ensuring the health and stability of river ecosystems, improving the scientificity and effectiveness of water resources management, promoting the coordinated development of economy, society and environment, and providing scientific basis for policy making. With the continuous in-depth research and widespread promotion of ecological regulation, ecological regulation will play a more important role in water resources management and river ecological protection. Therefore, the research on ecological regulation is particularly important.

[0003] At present, ecological regulation is a common and effective means of river ecological restoration. The traditional ecological regulation method of simulating flood peaks has the following problems:

[0004] (1) Traditional ecological flood peak simulation suffers from a lack of ecological data, such as a lack of early fish resource data and insufficient basic research, which limits the accurate simulation of flood processes;

[0005] (2) The simulation of ecological flood peaks for a few species is not very economical and scientific. The implementation of a certain ecological flood peak scheme may only benefit one or a few species. The scheduling that benefits more species is more meaningful. For the entire river ecosystem, it is obviously more meaningful to simulate the natural flood peak ecological scheduling to meet the scheduling of multiple species.

[0006] (3) Traditionally, the fitting of flood peak characteristic values ​​mostly uses the P-III curve and a single goodness of fit test method. This study considers the fitting of multiple probability density functions and sorts the goodness results by weight scores based on multiple goodness of fit tests, which increases the reliability and applicability of the results. Summary of the invention

[0007] In order to solve the current ecological scheduling scheme formulation, lack of ecological data, insufficient basic research, inaccurate simulation of flood processes, uneconomical and unscientific simulation of ecological scheduling flood peaks for a few species, and the single fitting curve and goodness test method of traditional flood peak characteristic values. The purpose of the present invention is to provide a reservoir ecological scheduling method that simulates natural flood peaks, which can perform multi-probability density function fitting on flood peak characteristic values ​​to solve the optimal flood peak characteristic values, and solve the flood peak characteristic value threshold based on the guarantee rate, and then determine the composition of the artificial flood peak process. The present invention increases the number of fish protection targets in the river ecosystem and more objectively reflects the needs of the river ecosystem, providing a necessary reference for the joint ecological scheduling of reservoir groups.

[0008] The technical solution adopted by the present invention is: a reservoir ecological dispatching method for simulating natural flood peaks, fitting flood peak characteristic values ​​through multiple probability density functions, calculating the optimal flood peak characteristic values ​​based on multiple superiority test methods, and solving the flood peak characteristic value threshold based on the guarantee rate, including the following steps:

[0009] S1. Collect data on hydrology, water temperature, fish resources and early fish resources in the study area;

[0010] S2. Determine the approximate location of fish spawning grounds in the study area, the approximate time of fish reproduction, and the approximate water temperature range for reproduction;

[0011] S3. Use multivariate diagnostic methods to analyze mutation points in multi-year runoff series;

[0012] S4. Analyze the peak characteristic values ​​of natural runoff sequence, and count the peak starting date, starting flow, rising time, rising amplitude, rising frequency, peak flow, precipitation time, precipitation amplitude, and characteristic value sequence of flow at the end of precipitation in typical fish breeding period in each year in different years of wet, normal and dry seasons;

[0013] S5. Perform multi-probability density function fitting analysis based on each multi-year flood peak characteristic value sequence;

[0014] S6. Perform goodness of fit analysis on each distribution function to determine the most suitable distribution function for each eigenvalue;

[0015] S7, solving the optimal flood peak characteristic values;

[0016] S8, using each multi-year flood peak characteristic value sequence to sort the frequency, and using a certain guarantee rate to calculate the threshold range of each flood peak characteristic value;

[0017] S9. Formulate reservoir ecological regulation plans for wet, normal and dry years;

[0018] S10. Verification and optimization of reservoir ecological dispatching scheme.

[0019] Furthermore, the step S3 specifically includes:

[0020] Based on the long-sequence daily flow data of the study area, a multi-variation diagnostic method was used to perform mutation point analysis to determine the variation year. The multi-variation diagnostic methods included cumulative deviation method, Man-Kendall method, Pettitt test method, Buishand U test method, Bayes test method, ordered clustering method, sliding F test method, Lee-Heghinan method, sliding T test method, sliding rank sum test method, and sliding runs test method. The year with the most jump points using multiple methods was the variation year. If the variation was significant, the long-sequence flow sequence before the variation was used as the natural runoff for the next analysis. If the variation was not significant, the long-sequence flow sequence before the dam was built was used as the natural runoff for the next analysis.

[0021] Furthermore, the step S5 specifically includes:

[0022] Based on the characteristic value sequences of flood peak rising date, rising flow, rising time, rising amplitude, rising frequency, precipitation time, precipitation amplitude and precipitation terminal flow, multiple probability density functions are fitted to each characteristic value sequence, and the multiple probability density functions include Gen.Extreme Value, Log-Logistic (3P), Gamma (3P), Log-Pearson 3, Gamma, Erlang, Log-Gamma, Nakagami, Erlang (3P), Inv.Gaussian, Normal, Inv.Gaussian (3P), Pearson 5 (3P), Fatigue Life (3P), Lognormal (3P), Pearson 5, Gen.Gamma, JohnsonSB, Lognormal, Cauchy, Fatigue Life, Error, Pearson 6, Logistic, Rayleigh (2P),

[0023] Chi-Squared(2P), Rice, Weibull(3P), Weibull, Hypersecant, Gumbel Max, Frechet(3P), Uniform, Log-Logistic, Gen.Pareto, Power Function,

[0024] Gumbel Min, Laplace, Levy, Levy(2P), Triangular, Frechet,

[0025] Beta, Reciprocal, Pareto, Exponential(2P), Kumaraswamy, Pert, Rayleigh, Exponential, Pareto 2, Chi-Squared, Dagum, Student's t, Burr, Burr(4P), Dagum(4P), Error Function, Gen.Gamma(4P), Pearson 6(4P).

[0026] Furthermore, the step S6 specifically includes:

[0027] The Kolmogorov-Smirnov (KS), Anderson-Darling (AD) and Chi-Squared (CS) test methods were used to perform goodness of fit test analysis to determine the optimal probability density function;

[0028] First, use the KS, AD and CS test values ​​to sort each distribution function from small to large as 1, 2, 3, ......., n, then the KS, AD and CS test values ​​in the ith distribution function are ranked K i , A i and C i ; If a test value does not exist, the distribution function is eliminated;

[0029] For the sort value K i , A i and C i Re-weight the sorting. The three tests are equally important. The weights are all set to 1, and there is a new sorting value S i for:

[0030] S i =K i+ A i+ C i (1)

[0031] The distribution functions are sorted according to the new sort value S i Sort from small to large, with S i The distribution function that is the smallest and has an extreme value is the optimal distribution function, thus determining the optimal probability density function.

[0032] Furthermore, the step S7 specifically includes:

[0033] The eigenvalue corresponding to the highest point of the optimal probability density function is the optimal eigenvalue. If there are multiple optimal probability density functions, the optimal eigenvalue is the average of the eigenvalues ​​corresponding to the highest points of multiple optimal probability density functions.

[0034] Furthermore, the step S8 specifically includes:

[0035] Based on the statistics of the peak flood start date, start flow, flood time, flood amplitude, flood frequency, peak flow, precipitation time, precipitation amplitude, and flow characteristic values ​​of the end of precipitation in typical fish breeding seasons in each year, frequency sorting is performed, and a certain guarantee rate is selected to calculate the threshold range of each peak characteristic value, including the upper threshold using the characteristic value sequence 25% guarantee rate, and the lower threshold using the characteristic value sequence 75% guarantee rate. If the most suitable characteristic value is outside the threshold range determined by the guarantee rate, the most suitable characteristic value is used as the upper or lower threshold.

[0036] Furthermore, the step S9 specifically includes:

[0037] In flood, normal and dry years, under the premise that the temperature is met, the artificial flood peak plan is formulated with reference to the most suitable flood rise date, according to the most suitable flood peak characteristic value, the most suitable starting flow, the most suitable flood rise amplitude, the most suitable flood rise time, the most suitable precipitation amplitude, the most suitable precipitation time and the most suitable precipitation flow at the end of precipitation. The threshold values ​​of each characteristic value constitute the artificial flood peak process.

[0038] Furthermore, the step S10 specifically includes:

[0039] Typical years of wet and dry seasons are selected for satisfaction analysis, and the artificial flood peak scheme is verified and optimized based on the satisfaction degree.

[0040] From the above, the above steps S4, S5, and S6 are key steps: (1) multiple probability density functions are used to fit the flood characteristic value sequence. There are differences in the optimal distribution functions of different basins and different flood characteristic values, which makes up for the inaccurate fitting of a single probability density function; (2) multiple goodness of fit test methods are used to comprehensively test the goodness of fit of the probability distribution function. The effects of a certain distribution function under different goodness of fit test methods are different. Based on multiple goodness of fit test methods, the present invention adopts a method of taking the distribution with the lowest comprehensive ranking score as the optimal distribution by weighted processing, which has strong scientific significance and applicability.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] (1) The present invention can solve the problems of lack of ecological data, such as lack of early fish resource data and insufficient basic research, in traditional ecological dispatch flood peak simulation, and realize accurate simulation of flood process;

[0043] (2) The present invention solves the problem that the traditional fitting of flood peak characteristic values ​​mostly adopts P-III curve, which cannot accurately fit the flood characteristic values, and the problem that a single goodness test method is insufficient. The present invention considers the fitting of multiple probability density functions, and sorts the goodness results by weighted scores based on multiple goodness tests, thereby increasing the reliability and applicability of the results;

[0044] (3) The present invention can perform multi-probability density function fitting on the flood peak characteristic value to solve the optimal flood peak characteristic value, and solve the flood peak characteristic value threshold based on the guarantee rate, thereby forming an artificial flood peak process and providing an ecological scheduling threshold space. Moreover, the flood peak simulation of the present invention is an ecological scheduling scheme that is beneficial to more species. For the entire river ecosystem, the solution of simulating the ecological scheduling of natural flood peaks is a solution that meets the ecological scheduling needs of multiple species, which is obviously more meaningful. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of an embodiment provided by the present invention.

[0046] Figure 2 It is a schematic diagram of multi-probability density function fitting of a flood characteristic value sequence provided by the present invention (the flow rate starting from May in a flood year), wherein the abscissa represents the flood peak characteristic value.

[0047] Figure 3 This is a schematic diagram of artificial flood scheduling in a river study area (in a normal water year in May). DETAILED DESCRIPTION

[0048] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. It should be noted that although the description of these embodiments is used to help understand the present invention, it does not constitute a limitation of the present invention. The specific structures and functional details disclosed herein are only used to describe the exemplary embodiments of the present invention. However, the present invention can be embodied in many alternative forms, and it should not be understood that the present invention is limited to the embodiments set forth herein.

[0049] A reservoir ecological dispatching method that simulates natural flood peaks, such as Figure 1 As shown, the following steps are included:

[0050] S1. Collect data on hydrology, water temperature, fish resources and early fish resources in the study area.

[0051] Fish ecological data are collected in the river section below the dam, and fish resources and early fish resources are analyzed. The fish ecological data include fish and egg and fry species, composition, development period, morphology, age structure, and ecological habits.

[0052] S2. Determine the approximate location of the fish spawning grounds, the approximate time of fish reproduction and the water temperature requirements.

[0053] The study area was determined based on historical fish spawning grounds, and a variety of representative drift-eating fish species were selected based on the historical spawning grounds and the dominance and ecological groups of drift-eating fish species.

[0054] The long-finned goby, long thin loach, short-bodied golden loach, etc. are important fish species that lay drifting eggs in a certain river section. The Chinese striped-breasted catfish, yellow stone climbing catfish, etc. are important fish species that lay sticky eggs in a certain river section. The ovulation and fertilization activities of mature broodstock require suitable water environment conditions and the signals and stimuli generated by their changes. Generally speaking, the spawning date is concentrated in May and June, and the water temperature is between 17 and 21 degrees Celsius.

[0055] S3. Use multi-variant diagnostic methods to analyze mutation points in multi-year runoff series.

[0056] Based on the flow data of a hydrological station in the study area from 1960 to 2023, the cumulative anomaly method, Man-Kendall method, ordered clustering method, sliding F test method, Lee-Heghinan method, sliding T test method, sliding rank sum test method, sliding run test method and other variation diagnosis methods are used to perform mutation point analysis to determine the variation year, and the year with the most jump points in multiple methods is the variation year. If the variation is significant, the long sequence flow sequence before the variation is used as the natural runoff for the next analysis; if the variation is not significant, the long sequence flow sequence before the dam is built is used as the natural runoff for the next analysis.

[0057] Table 1 Results of multivariate diagnostic methods

[0058]

[0059] As shown in Table 1, 11 common variation diagnosis methods were used to diagnose jump point variation, and it was found that the variation was not significant. Therefore, the flow sequence 1960-2008 before the construction of the regulating reservoir in a certain year upstream of a hydrological station was used for the next analysis.

[0060] S4. Analyze the characteristic values ​​of natural flood peaks, and divide the statistics into statistics of the starting date of flood, starting flow, flood time, flood amplitude, flood frequency, flood peak flow, precipitation time, precipitation amplitude, and characteristic values ​​of flow at the end of precipitation in flood, normal and dry years.

[0061] An analysis of the peak characteristic values ​​of natural runoff sequences was conducted, and the characteristic value sequences of the peak rising date, rising flow, rising time, rising amplitude, rising frequency, peak flow, precipitation time, precipitation amplitude, and flow at the end of precipitation during the typical fish breeding period were counted for flood, normal and dry years according to the guarantee rates of 25%, 25-75%, and 75%.

[0062] S5. Perform multi-probability density function fitting analysis based on multi-year flood peak characteristic value sequences.

[0063] Based on the characteristic value sequences of flood peak rising date, rising flow, rising time, rising amplitude, rising frequency, precipitation time, precipitation amplitude and precipitation terminal flow, multiple probability density functions are fitted to each characteristic value sequence. The multiple probability density functions include Gen.Extreme Value, Log-Logistic(3P), Gamma(3P), Log-Pearson 3, Gamma, Erlang, Log-Gamma, Nakagami, Erlang(3P), Inv.Gaussian, Normal, Inv.Gaussian(3P), Pearson 5(3P), Fatigue Life(3P), Lognormal(3P), Pearson 5, Gen.Gamma, Johnson SB, Lognormal, Cauchy, Fatigue Life, Error, Pearson 6, Logistic, Rayleigh(2P), Chi-Squared(2P), Rice, Weibull(3P), Weibull, Hypersecant, and Gumbel. Max, Frechet(3P),Uniform,Log-Logistic,Gen.Pareto,Power Function,Gumbel Min,Laplace,Levy,Levy(2P),Triangular,Frechet,Beta,Reciprocal,Pareto,Exponential(2P),Kumaraswamy,Pert,Rayleigh,Exponential,Pareto 2,Chi-Squared,Dagum,Student's t, Burr, Burr(4P), Dagum(4P), Error Function, Gen.Gamma(4P), Pearson 6(4P), such as Figure 2 shown.

[0064] S6. Perform goodness of fit analysis on each distribution function to determine the most suitable distribution function for each eigenvalue.

[0065] The Kolmogorov-Smirnov (KS), Anderson-Darling (AD) and Chi-Squared (CS) test methods were used to perform goodness of fit test analysis to determine the optimal probability density function.

[0066] First, use the KS, AD and CS test values ​​to sort each distribution function from small to large as 1, 2, 3, ..., n. Then the KS, AD and CS test values ​​in the ith distribution function are ranked K i , A i and C i If a test value does not exist, then remove the distribution function. Then sort the value K i , A i and C i Re-weight the sorting, assuming that the three tests are equally important, and set the weights to 1. Then there is a new sorting value S i for:

[0067] S i =K i+ A i+ C i (1)

[0068] The distribution functions are sorted according to the new sort value S i Sort from small to large, with S i The distribution function with the minimum and extreme value is the optimal distribution function, thus determining the optimal probability density function. As shown in Table 2, the optimal probability density function of the rising flow in May at a hydrological station is the Gen. Extreme Value distribution.

[0069] Table 2 Goodness of fit test table of 60 probability distribution functions of the rising flow in May at a hydrological station

[0070]

[0071]

[0072] S7. Solve for each optimal flood peak eigenvalue. The eigenvalue corresponding to the highest point of the optimal probability density function is the optimal eigenvalue. If there are multiple optimal probability density functions, the optimal eigenvalue is the average of the eigenvalues ​​corresponding to the highest points of multiple optimal probability density functions. As shown in Table 3.

[0073] Table 3 Optimal characteristic values ​​of flood peaks in May and June at a hydrological station

[0074]

[0075]

[0076] S8, using each multi-year flood peak characteristic value sequence to sort the frequency, and using a certain guarantee rate to calculate the threshold range of each flood peak characteristic value;

[0077] Based on the statistics of the peak flood start date, flood flow, flood time, flood amplitude, flood frequency, peak flow, precipitation time, precipitation amplitude, and precipitation end flow characteristic values ​​of typical fish breeding seasons in each year, frequency sorting is performed, and a certain guarantee rate is selected to calculate the threshold range of each peak characteristic value, such as the upper threshold using the characteristic value sequence 25% guarantee rate, and the lower threshold using the characteristic value sequence 75% guarantee rate. If the optimal characteristic value is outside the threshold range determined by the guarantee rate, the optimal characteristic value is used as the upper or lower threshold. As shown in Table 4.

[0078] Table 4 Thresholds of flood characteristic values ​​in May and June at a hydrological station

[0079]

[0080] S9. Formulate reservoir ecological dispatching plans for wet, normal and dry years.

[0081] In the years of flood, normal and dry seasons, under the premise of meeting the temperature requirements, the artificial flood peak plan is formulated with reference to the most suitable flood rise date, according to the most suitable flood peak characteristic value, the most suitable starting flow, the most suitable flood rise amplitude, the most suitable flood rise time, the most suitable precipitation amplitude, the most suitable precipitation time and the most suitable precipitation final flow. The threshold values ​​of each characteristic value constitute the artificial flood peak process (Table 5).

[0082] The artificial flood process in May of a normal water year starts on May 16 (according to the forecast, the water temperature on May 16 will be above 18°C, which meets the spawning water temperature requirements of most fish) (taking May of a normal water year as an example), and the starting flow is 890m 3 / s, the flood process lasted for 7 days (including rise and fall), of which the rising water lasted for 4 days and the falling water lasted for 3 days, and the peak flow was 1240m 3 / s, the average daily traffic increase is about 85m 3 / s(Daily average flow increase is 83~191m 3 / s), the average daily flow rate reduction is about 70m 3 / s(Daily average flow reduction value is 60~128m 3 / s), the average flow rate of the flood process is 1078m 3 / s. As shown in Table 5 and Figure 3 shown.

[0083] Table 5 Scheme of artificial flood dispatching process in a study area in May in a normal water year

[0084]

[0085] S10. Verification and optimization of reservoir ecological dispatching scheme.

[0086] Typical years of flood, normal and dry seasons were selected for satisfaction analysis, and the artificial flood peak scheme was verified and optimized according to the satisfaction degree (Table 6).

[0087] The reservoir operation status of a hydropower station in a typical year in May is shown in Table 6. In the flood year, an 8-day artificial flood ecological operation was carried out in mid-to-late May, with an average discharge of 1383m 3 / s. In normal and dry years, artificial flood ecological regulation is carried out for an average of 7 days, with average discharge flows of 1078 and 1008m 3 / s. The average discharge flow in the first and middle ten days of a dry year is less than 1008m 3 / s, while the average discharge in the second half of May is larger. The average discharge in the first half of May in a normal water year is 913m 3 / s, slightly less than 1078m 3 / s, but the average discharge in mid-May and late May of normal water years is 1142m 3 / s and 1735m 3 / s, significantly greater than 1078m 3 / s, which can be regulated by delaying the reservoir water storage for a few days. In the first, middle and last ten days of the flood year, the average outflow from the reservoir is 1391, 1414 and 1757m respectively. 3 / s, all greater than 1383m 3 / s. Moreover, the outflow from May 16 to 23 in a dry year, a normal year, and May 16 to 24 in a wet year were 1021, 1398, and 1504 m respectively. 3 / s, all greater than 1008, 1078 and 1383m 3 / s. Therefore, from the perspective of reservoir operation and water balance, it is feasible to implement the proposed artificial flood ecological operation in mid-to-late May from May 16 to 24, which will basically not affect the operation of reservoir water level in May and have little impact on the reservoir's annual operation process. As shown in Table 6.

[0088] Table 6 Reservoir dispatching and operation of a hydropower station in each typical year (May)

[0089]

[0090] What is disclosed above is only a preferred embodiment of the present invention, and cannot be used to limit the scope of rights of the present invention. Therefore, the present invention is not limited to the above optional implementation mode. Anyone can derive other various forms of products under the inspiration of the present invention. However, no matter what changes are made in its shape or structure, all technical solutions that fall within the scope defined by the claims of the present invention fall within the protection scope of the present invention.

Claims

1. A reservoir ecological dispatching method for simulating natural flood peaks, characterized in that: The following steps are involved: S1. Collect data on hydrology, water temperature, fish resources and early fish resources in the study area; S2. Determine the location of fish spawning grounds, fish breeding time and water temperature requirements; S3. Use multivariate diagnostic methods to analyze mutation points in multi-year runoff series; S4. Analyze the peak characteristic values ​​of natural runoff sequence, and count the peak starting date, starting flow, rising time, rising amplitude, rising frequency, peak flow, precipitation time, precipitation amplitude, and characteristic value sequence of flow at the end of precipitation in typical fish breeding period in each year in different years of wet, normal and dry seasons; S5. Perform multi-probability density function fitting analysis based on each multi-year flood peak characteristic value sequence; S6. Perform goodness of fit analysis on each distribution function to determine the most suitable distribution function for each eigenvalue; S7, solving the optimal flood peak characteristic values; S8, using each multi-year flood peak characteristic value sequence to sort the frequency, and using a certain guarantee rate to calculate the threshold range of each flood peak characteristic value; S9. Formulate reservoir ecological regulation plans for wet, normal and dry years; S10. Verification and optimization of reservoir ecological dispatching scheme.

2. The method according to claim 1, characterized in that The step S3 specifically includes: Based on the long-sequence daily flow data of the study area, a multi-variation diagnostic method is used to analyze the mutation points and determine the mutation year. The multi-variation diagnostic method includes cumulative deviation method, Man-Kendall method, Pettitt test method, BuishandU test method, Bayes test method, ordered clustering method, sliding F test method, Lee-Heghinan method, sliding T test method, sliding rank sum test method, and sliding run test method. The year with the most jump points in multiple methods is the mutation year. If the variation is significant, the long sequence flow sequence before the variation is used as the natural runoff for the next analysis; if the variation is not significant, the long sequence flow sequence before the dam is built is used as the natural runoff for the next analysis.

3. The method according to claim 1, characterized in that The step S5 specifically includes: Based on the characteristic value sequences of flood peak rising date, rising flow, rising time, rising amplitude, rising frequency, precipitation time, precipitation amplitude, and precipitation end flow, multiple probability density functions are fitted to each characteristic value sequence; The multiple probability density functions include Gen.Extreme Value, Log-Logistic(3P), Gamma(3P), Log-Pearson 3, Gamma, Erlang, Log-Gamma, Nakagami, Erlang(3P), Inv.Gaussian, Normal, Inv.Gaussian(3P), Pearson 5(3P), Fatigue Life(3P), Lognormal(3P), Pearson 5. Gen.Gamma, Johnson SB, Lognormal, Cauchy, Fatigue Life, Error, Pearson 6, Logistic, Rayleigh(2P), Chi-Squared(2P), Rice, Weibull(3P), Weibull, Hypersecant, Gumbel Max, Frechet(3P), Uniform, Log-Logistic, Gen.Pareto, Power Function, Gumbel Min, Laplace, Levy, Levy(2P), Triangular, Frechet, Beta, Reciprocal, Pareto, Exponential(2P), Kumaraswamy, Pert, Rayleigh, Exponential, Pareto 2, Chi-Squared, Dagum, Student's t, Burr, Burr(4P), Dagum(4P), Error Function, Gen.Gamma(4P), Pearson 6(4P).

4. The method according to claim 1, characterized in that The step S6 specifically includes: The Kolmogorov-Smirnov, Anderson-Darling and Chi-Squared test methods were used to perform goodness of fit test analysis to determine the optimal probability density function.

5. The method according to claim 4, characterized in that The Kolmogorov-Smirnov, Anderson-Darling and Chi-Squared test values ​​were used to rank the distribution functions from small to large as 1, 2, 3, ......., n; Then the ranking values ​​of Kolmogorov-Smirnov, Anderson-Darling and Chi-Squared test values ​​in the ith distribution function are K i , A i and C i , if a certain test value does not exist, then the distribution function is eliminated; For the sort value K i , A i and C i Re-weight the sorting. The three tests are equally important. The weights are all set to 1, and there is a new sorting value S i for: S i =K i+ A i+ C i (1) The distribution functions are sorted according to the new sort value S i Sort from small to large, with S i The distribution function that is the smallest and has an extreme value is the optimal distribution function, thus determining the optimal probability density function.

6. The method according to claim 1, characterized in that In step S7, the optimal eigenvalue is the eigenvalue corresponding to the highest point of the optimal probability density function. If there are multiple optimal probability density functions, the optimal eigenvalue is the average of the eigenvalues ​​corresponding to the highest points of the multiple optimal probability density functions.

7. The method according to claim 1, characterized in that The step S8 specifically includes: Based on the statistics of the characteristic values ​​of flood peak starting date, starting flow, rising time, rising amplitude, rising frequency, peak flow, precipitation time, precipitation amplitude and flow at the end of precipitation during the typical fish breeding period in each year in flood, normal and dry years, frequency sorting is performed, and a certain guarantee rate is selected to calculate the threshold range of each flood peak characteristic value. If the optimal characteristic value is outside the threshold range determined by the guarantee rate, the optimal characteristic value is used as the upper or lower threshold.

8. The method according to claim 7, characterized in that The certain guarantee rate includes an upper threshold value selecting a 25% guarantee rate of the characteristic value sequence, and a lower threshold value selecting a 75% guarantee rate of the characteristic value sequence.

9. The method according to claim 1, characterized in that: The step S9 specifically includes: In flood, normal and dry years, under the premise that the temperature is met, the artificial flood peak plan is formulated with reference to the most suitable flood rise date according to the most suitable flood peak characteristic values, the most suitable starting flow, the most suitable flood rise amplitude, the most suitable flood rise time, the most suitable precipitation amplitude, the most suitable precipitation time and the most suitable precipitation flow at the end of precipitation. The thresholds of various characteristic values ​​constitute the artificial flood peak process.

10. The method according to claim 1, characterized in that The step S10 specifically includes: Typical years of wet and dry seasons are selected for satisfaction analysis, and the artificial flood peak scheme is verified and optimized based on the satisfaction degree.

Images