A method for obtaining the upper limit of pre-storage capacity of cascade reservoirs considering rainfall forecast information and its uncertainty
Through the risk quantification method based on rainfall forecast information, the problem of insufficient calculation of flood limit water level in a single reservoir is solved. Using rainfall information in a longer foreseeable period, iterative calculations are used to determine the upper limit of pre-water storage of the cascade reservoir group, realizing the reservoir group Improve the efficiency of flood resource utilization and the effectiveness of flood prevention and disaster reduction.
Patent Information
- Application Number
- CN202411270302.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-11
AI Technical Summary
In the prior art, the upper limit calculation of flood limit water level dynamic control domains is mostly concentrated in a single reservoir, and the rainfall forecast information for a longer foreseeable period is not fully utilized, resulting in the problem of water abandonment or no water sustainability in the flood season, and it has not been effectively improved. Flood resource utilization rate of cascade reservoir groups.
The risk quantification method based on rainfall forecast information is adopted, and the numerical rainfall forecast accuracy and uncertainty analysis are used to determine the upper limit of pre-water storage of the cascade reservoir group through cyclic iterative calculations to ensure that no additional flood control risks are increased and that the flood control standards of each reservoir are met.
The upper limit of pre-water storage capacity of the cascade reservoir group is improved without increasing flood control risks, and the process-based calculation method is improved, so as to improve the flood resource utilization efficiency and flood control and disaster reduction capabilities of the reservoir group.
Smart Images

Figure CN118798488B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of reservoir scheduling and relates to a method for obtaining an upper limit of pre-storage water volume of a cascade reservoir group taking rainfall forecast information and its uncertainty into consideration. Background Art
[0002] Flood resource utilization is an important measure to alleviate water shortage, achieve optimal allocation of water resources and sustainable development. Reservoirs are the most effective engineering measures for flood resource utilization in river basins. The flood limit water level is the key to balancing flood control risks and benefits for reservoirs. The scientific and reasonable value of the flood limit water level determines the common storage capacity of reservoirs for flood control and benefits, and provides operational space for improving the utilization rate of flood resources [Li Xiaoying, Zheng Haoran, Wu Shujun, et al. Analysis of the current status and potential of flood resource utilization in river basins based on binary limit theory [J]. Engineering Science and Technology, 2020, 52(02): 70-77.]. The traditional flood limit water level is formulated at the planning and design level based on the multi-year average water inflow characteristics of the flood season and the reservoir regulation capacity. A single value is used throughout the flood season, resulting in the situation in which reservoirs abandon water during the flood season and have no water to replenish after the flood season in actual operation. In addition, the traditional flood limit water level formulation only considers a single reservoir [A method for specifying dynamic flood limit water level of reservoirs, publication number CN 114239992 A], and the reservoir group coordination capability is not fully utilized. In order to improve the utilization rate of basin flood resources, the dynamic control of flood limit water level of reservoir groups has become one of the main research directions of reservoir flood resource utilization.
[0003] Dynamic control of flood-limited water level refers to the dynamic floating of water level in the dynamic control domain of flood-limited water level according to forecast information during reservoir operation, which can alleviate the contradiction between flood control and benefit development of reservoirs under the premise of ensuring flood control safety. The upper limit of pre-storage water volume (the amount of water between the upper and lower limits of the dynamic control domain of flood-limited water level) is the key parameter for dynamic control of flood-limited water level of reservoir groups, and its value is significantly affected by the length of the forecast information forecast period. With the development of hydrological and meteorological forecasting technology, the widespread and stable operation of water and rainfall monitoring and flood forecasting and dispatching systems, the accuracy and forecast period of forecast information have been effectively improved and extended, which provides a possibility for reservoirs to achieve dynamic control of flood limit water levels and further tap the potential of flood resource utilization [Tian Xingli, Huang Qiang, Wang Yimin, et al. Research progress on flood resource utilization of reservoirs in my country [J]. Journal of Water Resources and Water Engineering, 2011, 22(05): 24-28.], but existing research has not fully utilized rainfall forecast information to increase reservoir pre-storage [A method for determining dynamic flood limit water levels in flood seasons based on pre-discharge rules, publication number CN 114707879 B]. On the one hand, rainfall forecast information with a longer forecast period can leave reservoirs with more time to make dispatching decisions, thereby increasing the upper limit of pre-storage water. On the other hand, due to the uncertainty of rainfall forecast information, using rainfall forecast information to raise the flood limit water level may lead to certain flood control risks. The longer the forecast period, the greater the uncertainty of rainfall forecast information, and the higher the probability of risk after the reservoir flood limit water level is raised. Therefore, the increase in the flood limit water level needs to be constrained by not increasing the flood control risk. The risk event that may occur after the flood limit water level is raised is essentially the use of rainfall forecast information to raise the flood limit water level and encounter rainfall of different frequencies, causing the reservoir flood control high water level to exceed the characteristic water level. Therefore, the risk calculation method based on rainfall forecast information can be used to calculate the upper limit of the pre-storage capacity of the cascade reservoir group, with the maximum rainfall frequency of risk events as the risk rate, and the risk rate not exceeding the design risk rate as a constraint.
[0004] Each reservoir in the reservoir group system has the functions of hydrological compensation and storage capacity compensation. Therefore, compared with the dynamic control of the flood limit water level of a single reservoir, the joint dynamic control of the flood limit water level of the reservoir group has greater potential for flood resource utilization, and plays a more important role in improving the level of flood control and disaster reduction in the basin and the benefits of flood resource utilization [Zhou Xinchun, Xu Yinshan, Feng Baofei. Preliminary study on the interoperability of flood control storage capacity of the cascade reservoir group in the upper reaches of the Yangtze River [J]. Advances in Water Science, 2017, 28(03): 421-428.]. However, the hydraulic connection of the cascade reservoir group is complex, and there are many flood control targets. Each reservoir not only has its own flood protection object, but the reservoir group system also needs to jointly protect the downstream flood control safety, which brings challenges to the calculation of the upper limit of the pre-storage water volume of the reservoir group system. To this end, the present invention aims at the problem that the upper limit calculation of the existing flood limit water level dynamic control domain is mostly concentrated on a single reservoir, and the forecast period is short and rainfall forecast information is not considered. A method for calculating the upper limit of the pre-storage water volume of the cascade reservoir group using rainfall forecast information and risk analysis is proposed. This method uses numerical rainfall forecast information as the pre-storage evaluation index, and takes each reservoir as the constraint of not increasing the flood control risk. It quantifies the upper limit of pre-storage water that meets the flood control standards of cascade reservoirs under different forecast rainfall conditions. This method has certain universality in determining the upper limit of pre-storage water for cascade reservoirs. Summary of the invention
[0005] In view of the problem that the calculation of the upper limit of the dynamic control domain of the traditional flood limit water level is mostly concentrated on a single reservoir, and the forecast period is short and rainfall forecast information is not considered, the present invention provides a method for calculating the upper limit of the pre-storage water volume of a cascade reservoir group using rainfall forecast information and risk analysis. This method relies on forecast rainfall information and adopts a risk quantification method based on forecast rainfall information to calculate the flood control risk of reservoirs under different forecast rainfall information. Under the condition of ensuring that no additional flood control risk is added, the calculation is iterated cyclically to determine the upper limit of the pre-storage water volume of the reservoir group that meets the risk constraints. This method clarifies the calculation process of the upper limit of the pre-storage water volume of the reservoir group, and uses the Jinxia-Three Gorges cascade reservoir as an actual case to verify the rationality of the present invention.
[0006] To achieve the above object, the technical solution of the present invention is as follows:
[0007] A method for obtaining the upper limit of pre-storage water volume of a cascade reservoir group using rainfall forecast information and risk analysis. The method for calculating the upper limit of pre-storage water volume of a cascade reservoir group first analyzes the accuracy and availability of numerical rainfall forecast information, including graded numerical rainfall forecasts, and determines the available rainfall forecast information and forecast period. Then, the uncertainty of rainfall forecast information is analyzed, including statistics on actual rainfall under different rainfall levels, and determination of possible rainfall under different rainfall forecast levels and its corresponding probability set; secondly, the theoretical pre-storage water volume of each reservoir in the cascade reservoir group is calculated using the pre-discharge capacity constraint method based on aggregation-decomposition; thirdly, the flood process that may be formed under different forecast rainfall levels is deduced; finally, the risk calculation method based on rainfall forecast information is used to cyclically calculate the flood control risk of the reservoir group corresponding to different pre-storage water volume combinations, and the maximum pre-storage water volume of the original design is used as the upper limit of the pre-storage water volume of the reservoir group for flood control risks that do not exceed the original design. The details are as follows:
[0008] Step 1: Analyze the accuracy and availability of numerical rainfall forecast information
[0009] 1.1) The numerical rainfall forecasts are classified and then the accuracy of the rainfall forecasts at different levels, in different sub-basins and in different forecast periods is analyzed, including the accuracy rate, underreporting rate and false alarm rate, as shown in formulas (1) to (3).
[0010] ; (1)
[0011] ; (2)
[0012] ; (3)
[0013] Where: TS, FAR, PO are accuracy, missed alarm rate, and false alarm rate, respectively. A is the number of times the actual rainfall level is the same as the forecast rainfall level. B is the number of times the actual rainfall level is greater than the forecast rainfall level, N C It is the number of times the actual rainfall level is less than the forecast rainfall level.
[0014] 1.2) Based on the accuracy analysis of numerical rainfall forecast in step 1.1), the forecast period and basin evolution time are further compared to identify and select the available rainfall forecast information and forecast period.
[0015] Step 2: Analyze the uncertainty of numerical rainfall forecast information
[0016] The uncertainty of numerical rainfall forecast information can be represented by the actual rainfall distribution function under a given rainfall forecast level. In addition, there may be strong spatial correlation between the numerical rainfall forecast errors in different sub-basins. Therefore, it is necessary to select a suitable method to estimate the joint density function of the numerical rainfall forecast errors in different sub-basins. Most existing studies use P-Ⅲ distribution to derive the distribution function of hydrological elements. The following is a method for deriving the P-Ⅲ distribution function.
[0017] Step 2.1) Use the measured rainfall series to calculate the mean, dispersion coefficient and skewness coefficient of rainfall at different forecast levels. , coefficient of dispersion C v and the skewness coefficient C s As shown in formulas (4)-(7) respectively.
[0018] ; (4)
[0019] ; (5)
[0020] ; (6)
[0021] ; (7)
[0022] Where: x i is the rainfall sequence value, i=1,2,3,···,n; n is the total number of sequences. b is C s and C v The ratio of can be determined by the method of fitting; C s represents the skewness coefficient; C v Represents the coefficient of dispersion.
[0023] Step 2.2) Calculate the characteristic parameters of the actual rainfall distribution function, as shown in formulas (8) to (10).
[0024] ; (8)
[0025] ; (9)
[0026] ; (10)
[0027] Where: is the mean of the rainfall series data; , , are the parameters that characterize the shape, size, and position of the Pearson type III distribution, >0, >0.
[0028] Step 2.3) Calculate the probability density function of the actual rainfall distribution curve, as shown in formula (11).
[0029] ; (11)
[0030] Where: for The gamma function.
[0031] Step 2.4) Repeat the above steps to calculate the probability density function of the actual rainfall distribution curve of the m reservoirs in the basin .
[0032] Step 2.5) Estimate the joint density function of the numerical rainfall forecast error for each sub-basin, as shown in formula (12).
[0033] F ( x 1 , x 2 , ⋯ , x m ) = C [ f ( x 1 ), f ( x 2 ), ⋯ , f ( x m ) ] ; (12)
[0034] Where: is the calculation function of joint density distribution; represents the probability density function of the actual rainfall distribution in the basin controlled by the mth reservoir; m represents the number of cascade reservoirs.
[0035] Step 2.6) Propose different probability sets {p i |i=1,2,..,z}, use the 2.5 formula to convert p i Substitute the joint probability density function of the numerical rainfall forecast error to obtain the possible rainfall set {X(p i )|i=1,2,..,z}, where z represents the total number of elements in the probability set and rainfall set, and i represents the order of the elements in the set.
[0036] Step 3: Calculate the theoretical pre-storage capacity of each reservoir in the cascade reservoir group based on the aggregation-decomposition pre-discharge capacity constraint method
[0037] Step 3.1) Aggregate the cascade reservoirs into a virtual reservoir and calculate the maximum pre-storage capacity of the “aggregated reservoir” during the forecast period under the flood control safety conditions of the common protection objects as the theoretical pre-storage capacity of the “aggregated reservoir”, as shown in formulas (13) to (15).
[0038] ; (13)
[0039] Z d ' = f [ V ( Z d 0 ) + w ] ; (14)
[0040] ; (15)
[0041] Where: The theoretical pre-storage volume of the “aggregation reservoir”; Flood limit water level The corresponding storage capacity; To raise the flood limit water level after pre-storage; It is the upper limit value of dynamic control of flood limit water level; To store water for benefit; f [ * ] is the reservoir capacity-water level relationship; The safe flow for the common flood protection objects of the “aggregate reservoirs”; for Average inflow within for Average discharge flow within For the forecast period.
[0042] Step 3.2) According to the hydraulic connection of the cascade reservoirs and the flood control standards of the protected objects of each reservoir, the theoretical pre-storage water volume calculated in step 3.1 is coordinated and "decomposed" to each reservoir. In other words, the theoretical pre-storage water volume of each reservoir that meets the flood control standards is decomposed and calculated, as shown in formulas (16) to (18).
[0043] Z d , j + = f j [ V j ( Z d , j 0 ) + w j ] ; (16)
[0044] ; (17)
[0045] ; (18)
[0046] In the formula, Indicates the water level corresponding to the pre-storage capacity of the jth reservoir in the cascade reservoirs; f j [ * ] represents the storage capacity-water level relationship of the jth reservoir; represents the storage capacity corresponding to the original design flood limit water level of the jth reservoir; represents the pre-storage capacity of the jth reservoir. is the jth reservoir Average inflow within is the jth reservoir Average discharge flow within is the safe flow of the jth reservoir flood protection object.
[0047] Step 4: Determine the flood process that may be caused by different rainfall amounts under rainfall uncertainty. Select a typical flood process and calculate the net rainfall corresponding to the typical flood process and X(p i ) value corresponds to the ratio of net rainfall, and the typical flood process is magnified by the same ratio to obtain X(p i ) value of flood process SF(pi ).
[0048] Step 5: Taking the theoretical pre-storage volume obtained in step 3.1) as the boundary, the risk quantification method based on rainfall forecast information is used to calculate the flood control risk of each reservoir under different pre-storage volume combinations, and the maximum pre-storage volume of each reservoir that does not exceed the original design is used as the upper limit of the reservoir pre-storage volume. The risk quantification formula and risk constraint based on rainfall forecast information are shown in formula (18) and formula (19):
[0049] P j , k ( Z d , j , Z m , i ) = P Z { Z [ Z d , j , X ( p i ) ] ≥ Z m , i } ; (19)
[0050] ; (20)
[0051] In the formula, Z [ Z d , j , X ( p i ) ] is the jth reservoir To adjust the water level, adjust The high water level of flood caused by rainfall error; is the original design risk rate of the jth reservoir; P Z The rainfall frequency corresponding to the original designed flood control high water level is calculated in the flood control high water level-rainfall frequency curve.
[0052] The specific implementation steps of reservoir flood control risks under different pre-storage water volumes and the upper limit of pre-storage water volume that meets risk constraints are as follows:
[0053] Step 5.1) Let j=1.
[0054] Step 5.2) The theoretical pre-storage capacity of the jth reservoir is uniformly discretized, and the theoretical pre-storage capacity of the reservoir is used as the boundary to obtain the pre-storage capacity set of the reservoir {w j,k |j=1,2,..,m;k=1,2,..,l}, l is the number of elements in the pre-storage water volume set.
[0055] Step 5.3) Let k=1.
[0056] Step 5.4) The reservoir is pre-stored with water volume w j,k Corresponding water level Z d,j When adjusting the rainfall forecast in the conventional scheduling mode, a rainfall frequency p i The flood process SF(p i ), and obtain the corresponding flood control high water level Z [ Z d , j , X ( p i ) ] ; For each rainfall frequency {(p i )|i=1,2,..,z} corresponds to the flood process {SF(p i )|i=1,2,..,z} is used for regulation calculation to obtain the flood regulation high water level-rainfall frequency curve.
[0057] Step 5.5) On the flood control high water level-rainfall frequency curve, interpolate to obtain the original designed flood control high water level Z m,j The corresponding frequency P Z , this frequency is the pre-storage water volume w j,k The corresponding flood risk rate P j,k .
[0058] Step 5.6) If P j,k Less than P d,j , then let k=k+1 and repeat steps 5.4 and 5.5. j,k Greater than P d,j , then the upper limit of the reservoir pre-storage capacity is w j,k-1 .
[0059] Step 5.7) Let j = j + 1, and repeat steps 5.2 to 5.6 until the upper limits of pre-storage capacity of all reservoirs are calculated, and the upper limits of pre-storage capacity of the cascade reservoir group are obtained.
[0060] The above-mentioned method for calculating the upper limit of pre-storage water volume of cascade reservoirs using rainfall forecast information and risk analysis is applied to reservoir scheduling technology.
[0061] The beneficial effects of the present invention are:
[0062] Aiming at the problem that the calculation of the upper limit of the dynamic control domain of the traditional flood limit water level is mostly concentrated on a single reservoir and does not use the forecast information of a longer forecast period. The present invention constructs a method for calculating the upper limit of the pre-storage water volume of a cascade reservoir group using rainfall forecast information and risk analysis. Based on the forecast rainfall information of a longer forecast period, the risk quantification method based on rainfall forecast information is used to determine the upper limit of the pre-storage water volume of each reservoir in the reservoir group system without increasing the flood control risk. The process-based calculation method of the upper limit of the pre-storage water volume of the reservoir group meets the needs of reservoir flood resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS
[0063] Figure 1 It is a schematic diagram of the flow chart of calculating the upper limit of the pre-storage water volume of a cascade reservoir group using rainfall forecast information and risk analysis in the present invention;
[0064] Figure 2 It is a numerical rainfall forecast accuracy map of different levels for different forecast periods in each sub-basin in the example application of the present invention; among them, (a) is a numerical rainfall forecast accuracy map of L1 level for different forecast periods, and (b) is a numerical rainfall forecast accuracy map of L2 level for different forecast periods.
[0065] Figure 3It is a distribution curve diagram of actual rainfall at different forecast rainfall levels in a 3-day forecast period in an example application of the present invention, wherein Figure (a) is a distribution curve diagram of actual rainfall at L1 level forecast rainfall, and Figure (b) is a distribution curve diagram of actual rainfall at L2 level forecast rainfall.
[0066] Figure 4 This is a graph showing changes in the upper limit of the pre-storage water volume of the Three Gorges Reservoir with the forecast period in a cascade reservoir group system in an example application of the present invention. DETAILED DESCRIPTION
[0067] The present invention is further described below in conjunction with specific embodiments.
[0068] The present invention proposes a method for calculating the upper limit of pre-storage water volume of cascade reservoirs using rainfall forecast information and risk analysis. The following takes the Three Gorges cascade reservoir group in the lower reaches of the Jinsha River as a research example, and uses the ECMWF rainfall forecast product from May to October 2014 to 2021 as forecast information to calculate the upper limit of pre-storage water volume of cascade reservoirs. The main steps are as follows:
[0069] Step 1: Analyze the accuracy and availability of numerical rainfall forecast information
[0070] Step 1.1) Since the study basin area is large, the spatial heterogeneity of rainfall is averaged, resulting in a small number of light rain and heavy rain samples in the basin. Therefore, in this example, the five rainfall levels are merged into three, namely L1 (no rain, light rain), L2 (moderate rain) and L3 (heavy rain, heavy rain). According to formulas (1) to (3), the accuracy of numerical rainfall forecasts for different basins, different levels and different forecast periods is calculated. The average of each sub-basin is as follows: Figure 2 shown.
[0071] Step 1.2) Based on the accuracy analysis of numerical rainfall forecast, it can be seen that the L1 and L2 level rainfall forecasts in this example are highly accurate and can be used for pre-storage water calculation. At the same time, the evolution time from the four reservoirs in the lower reaches of the Jinsha River to the Three Gorges Reservoir is about 2 days, so the forecast period of numerical rainfall forecasts that can be used in this example is at least 3 days.
[0072] Step 2: Analyze the uncertainty of numerical rainfall forecast information
[0073] Based on the measured rainfall data, the probability density function of the actual rainfall distribution curve of each sub-basin is calculated using formulas (4) to (12), and then the joint density function of the numerical rainfall forecast error of each sub-basin is derived. The 3D forecast period joint density function curve is as follows: Figure 3 shown.
[0074] Given different probability sets {0.000001%, 0.0001%, 0.01%, 0.1%, 1%, 2%, 5%, 10%, 20%, 50%, 99%}. Substituting the probability set into the curve, we can get different forecast rainfall levels and different probabilities {X(P i )|i=1,2,..,m}, the possible rainfall amount X(P i ), and the results are shown in Table 1.
[0075] Table 1 Possible rainfall with different probabilities under different forecast rainfall levels
[0076]
[0077] Step 3: Calculate the theoretical pre-storage capacity of each reservoir in the reservoir group using the pre-discharge capacity constraint method based on aggregate decomposition
[0078] Step 3.1) According to the measured data, the average inflow of the reservoir system is 20000m 3 / s, the overall downstream safety discharge of the system is 35000m 3 / s, the pre-discharge time of the 3-7D forecast period needs to deduct 2h of operation time. Substituting the above parameters into formula (10), the theoretical pre-storage capacity of the Three Gorges Reservoir and Jinxia cascade reservoirs in different forecast periods can be obtained, as shown in Table 2.
[0079] Table 2 Theoretical pre-storage capacity of reservoir system at different forecast periods
[0080]
[0081] Step 3.2) The average inflow of the lower Jin cascade is 3300m 3 / s, the safe discharge volume downstream of Jinxia cascade is 12000m 3 / s, and the parameters are substituted into formula (12) to decompose the overall pre-storage capacity of the reservoir group into two parts: the Three Gorges and the Jinxia cascade, as shown in Table 3.
[0082] Table 3 Theoretical pre-storage water volume of the Three Gorges and Jinxia cascades at different forecast periods
[0083]
[0084] Step 4: Determine the flood process that may be caused by different rainfall amounts under rainfall uncertainty.
[0085] The peak flood level of the Three Gorges Reservoir in mid-September 2020 was 39,000 m 3 / s small and medium floods are taken as typical examples, and the net rainfall corresponding to the typical process is calculated and the X(p i ) value corresponds to the ratio of net rainfall, and the typical flood process is magnified by the same ratio to obtain X(p i) value of flood process SF(p i ).
[0086] Step 5: Taking the theoretical pre-storage water volume obtained in step 3.1) as the boundary, use the risk quantification method based on rainfall forecast information to calculate the flood control risk of each reservoir under different pre-storage water volume combinations, and take the maximum pre-storage water volume of each reservoir that does not exceed the original design as the upper limit of the reservoir pre-storage water volume.
[0087] The specific implementation steps of reservoir flood control risks under different pre-storage water volumes and the upper limit of pre-storage water volume that meets risk constraints are as follows:
[0088] Step 5.1) Let j=1.
[0089] Step 5.2) The theoretical pre-storage capacity of the jth reservoir is uniformly discretized, and the theoretical pre-storage capacity of the reservoir is used as the boundary to obtain the pre-storage capacity set of the reservoir {w j,k |j=1,2,..,m;k=1,2,..,l}, l is the number of elements in the pre-storage water set. For example, the pre-storage water set for the Three Gorges Reservoir in the 3D forecast period is {0,1,2,…,37}.
[0090] Step 5.3) Let k=1.
[0091] Step 5.4) The reservoir is pre-stored with water volume w j,k Corresponding water level Z d,j When adjusting the rainfall forecast in the conventional scheduling mode, a rainfall frequency p i The flood process SF(p i ), and obtain the corresponding flood control high water level Z [ Z d , j , X ( p i ) ] ; For each rainfall frequency {(p i )|i=1,2,..,z} corresponds to the flood process {SF(p i )|i=1,2,..,z} to calculate the flood regulation high water level-rainfall frequency curve. Table 4 shows the flood regulation high water level of the Three Gorges Reservoir under different forecast levels and rainfall frequencies when the pre-storage water volume is 0.
[0092] Table 4 High water level of Three Gorges flood control
[0093]
[0094] Step 5.5) On the flood control high water level-rainfall frequency curve, interpolate to obtain the original designed flood control high water level Z m,j The corresponding frequency P Z , this frequency is the pre-storage water volume w j,k The corresponding flood risk rate P j,k Table 5 gives the flood risk rates corresponding to different pre-storage volumes of the Three Gorges Reservoir under the L1 level forecast rainfall conditions in the 3D forecast period.
[0095] Table 5 High water level of the Three Gorges flood control
[0096]
[0097] Step 5.6) If P j,k Less than P d,j , then let k=k+1 and repeat steps 5.4 and 5.5. j,k Greater than P d,j , then the upper limit of the reservoir pre-storage capacity is w j,k-1 As shown in Table 5, the pre-storage water volume is 2.1 billion m 3 At that time, the risk rate of the Three Gorges Reservoir was 21.35%, 20% higher than the designed risk rate, and the pre-storage water volume was 2 billion m 3 The risk rate of the Three Gorges Reservoir was 19.71%, which was 20% lower than the designed risk rate. Therefore, the water storage capacity of the Three Gorges Reservoir in Shaanxi Province was 2 billion m3 when the 3D forecast period L1 level rainfall forecast was used. 3 .
[0098] Step 5.7) Let j = j + 1, and repeat steps 5.2 to 5.6 until the upper limits of pre-storage capacity of all reservoirs are calculated, and the upper limits of pre-storage capacity of the cascade reservoir group are obtained. Figure 4 This is a graph showing the change in the upper limit of the Three Gorges Reservoir's pre-storage water volume over the forecast period.
[0099] The above results show that the method for calculating the upper limit of pre-storage water volume of cascade reservoir groups proposed in the present invention can rely on rainfall forecast information with a longer foreseeable period to determine the upper limit of pre-storage water volume of each reservoir in the cascade reservoir group system without increasing the flood control risk. The method for calculating the upper limit of pre-storage water volume of cascade reservoir groups can be streamlined to meet the needs of flood resource utilization of cascade reservoir groups.
[0100] The above-described embodiments only represent a specific implementation of the present invention and cannot be regarded as limiting the scope of the patent right of the present invention. It should be emphasized that technical professionals in this field may still make various modifications and improvements without departing from the basic principles of the present invention, which are all included in the scope of the patent right of the present invention.
Claims
1. A method for obtaining the upper limit of pre-storage water volume of a cascade reservoir group considering rainfall forecast information and its uncertainty, characterized in that: Firstly, the accuracy and availability of numerical rainfall forecast information are analyzed, including graded numerical rainfall forecast, to determine the available rainfall forecast information and forecast period; then, the uncertainty of rainfall forecast information is analyzed, including statistics of actual rainfall under different rainfall levels, to determine the possible rainfall under different rainfall forecast levels and their corresponding probability sets; secondly, the theoretical pre-storage capacity of each reservoir in the cascade reservoir group is calculated using the pre-discharge capacity constraint method based on aggregation-decomposition; Secondly, the flood process that may be formed under different forecast rainfall levels is deduced; finally, the risk calculation method based on rainfall forecast information is used to cyclically calculate the flood control risk of the reservoir group corresponding to different pre-storage water volume combinations, and the maximum pre-storage water volume of the reservoir group is taken as the upper limit of the flood control risk not exceeding the original design; The specific steps are as follows: Step 1: Analyze the accuracy and availability of numerical rainfall forecast information, including classifying numerical rainfall forecasts, and determine the available rainfall forecast information and forecast period; Step 2: Analyze the uncertainty of numerical rainfall forecast information The uncertainty of numerical rainfall forecast information is represented by the actual rainfall distribution function under a given rainfall forecast level, and the joint density function of numerical rainfall forecast errors in different sub-basins is estimated; the distribution function of hydrological elements is derived using the P-III distribution; the derivation method of the P-III distribution function is as follows: Step 2.1) Use the measured rainfall series to calculate the mean, deviation coefficient and skewness coefficient of rainfall at different forecast levels. The mean x, deviation coefficient C v and the skewness coefficient C s As shown in formulas (4)-(7) respectively; C s =bC v (7) Where: x i is the rainfall sequence value, i=1,2,3,···,n; n is the total number of sequences; b is C s and C v The ratio of is determined by the method of fitting; C s represents the skewness coefficient; C v represents the coefficient of dispersion; Step 2.2) Calculate the characteristic parameters of the actual rainfall distribution function, as shown in formula (8) to formula (10); Where: is the mean of the rainfall series data; α, β, and a0 are the parameters that characterize the shape, size, and location of the Pearson III distribution, respectively, α>0, β>0; Step 2.3) Calculate the probability density function of the actual rainfall distribution curve, as shown in formula (11); Where: Γ(α) is the gamma function of α; Step 2.4) Repeat the above steps to calculate the probability density function f(x1), f(x2),…, f(x m ); 2.5) Estimate the joint density function of the numerical rainfall forecast error of each sub-basin, as shown in formula (12); F(x1,x2,…,x m )=C[f(x1),f(x2),…,f(x m )];(12) Where: C[*] is the calculation function of the joint density distribution; f(x m ) represents the probability density function of the actual rainfall distribution in the basin controlled by the mth reservoir; m represents the number of cascade reservoirs; Step 2.6) Prepare different probability sets {p i |i=1,2,..,z}, use the 2.5 formula to convert p i Substitute the joint probability density function of the numerical rainfall forecast error to obtain the possible rainfall set {X(p i )|i=1,2,..,z}, where z represents the total number of elements in the probability set and rainfall set, and i represents the order of the elements in the set; Step 3: Calculate the theoretical pre-storage capacity of each reservoir in the cascade reservoir group based on the aggregation-decomposition pre-discharge capacity constraint method Step 3.1) Aggregate the cascade reservoirs into a virtual reservoir, and calculate the maximum pre-storage capacity of the "aggregate reservoir" during the forecast period under the flood control safety conditions of the common protection object as the theoretical pre-storage capacity of the "aggregate reservoir", as shown in formulas (13) to (15); w=(Q out -Q in )T y ,Q out ≤Q an ;(13) Where: w is the theoretical pre-storage water volume of the "aggregate reservoir"; Flood limit water level Corresponding storage capacity; Z' d To raise the flood limit water level after pre-storage; is the upper limit of the dynamic control of the flood limit water level; Z XL is the beneficial water storage level; f[*] is the relationship between storage capacity and water level; Q an The safe flow rate of the "aggregate reservoir" for common flood protection; Q in T y The average inflow within out T y Average discharge flow in y For the foreseeable period; Step 3.2) The theoretical pre-storage water volume calculated in step 3.1) is coordinated and "decomposed" to each reservoir according to the hydraulic connection of the cascade reservoir group and the flood control standard of each reservoir protection object; that is, the theoretical pre-storage water volume of each reservoir that meets the flood control standard is decomposed and calculated, as shown in formulas (16) to (18); w j =(Q out,j -Q in,j )T y ,Q out,j ≤Q an,j ;(17) In the formula, represents the water level corresponding to the pre-storage capacity of the jth reservoir in the cascade reservoirs; f j [*] represents the storage capacity-water level relationship of the jth reservoir; represents the storage capacity corresponding to the original design flood limit water level of the jth reservoir; w yx,j represents the pre-storage capacity of the jth reservoir; Q in,j is the jth reservoir T y The average inflow within out,j is the jth reservoir T y Average leakage flow in an,j is the safe flow of the jth reservoir flood protection object; Step 4: Select a typical flood process and calculate the net rainfall corresponding to the typical flood process and X(p i ) value corresponds to the ratio of net rainfall, and the typical flood process is magnified by the same ratio to obtain X(p i ) value of flood process SF(p i ); Step 5: Taking the theoretical pre-storage water volume obtained in step 3.1) as the boundary, the risk quantification method based on rainfall forecast information is used to calculate the flood control risk of each reservoir under different pre-storage water volume combinations, and the maximum pre-storage water volume of each reservoir that does not exceed the original design is used as the upper limit of the reservoir pre-storage water volume; the specific implementation steps of the reservoir flood control risk under different pre-storage water volumes and the upper limit of pre-storage water volume that meets the risk constraints are as follows: Step 5.1) Let j = 1; Step 5.2) The theoretical pre-storage capacity of the jth reservoir is uniformly discretized, and the theoretical pre-storage capacity of the reservoir is used as the boundary to obtain the pre-storage capacity set {w j,k |j=1,2,..,m;k=1,2,..,l}, l is the number of elements in the pre-storage water volume set; Step 5.3) Set k=1; Step 5.4) The reservoir is pre-stored with water volume w j,k Corresponding water level Z d,j When adjusting the rainfall forecast in the conventional scheduling mode, a rainfall frequency p i The flood process SF(p i ), and obtain the corresponding flood control high water level Z[Z d,j ,X(p i )]; for each rainfall frequency {(p i )|i=1,2,..,z}corresponding to the flood process {SF(p i )|i=1,2,..,z} to perform regulation calculation and obtain the flood regulation high water level-rainfall frequency curve; Step 5.5) On the flood control high water level-rainfall frequency curve, interpolate the original designed flood control high water level Z m,j The corresponding frequency P Z , this frequency is the pre-storage water volume w j,k The corresponding flood risk rate P j,k ; Step 5.6) If P j,k Less than P d,j , then let k = k + 1, repeat steps 5.4 and 5.5; if P j,k Greater than P d,j , then the upper limit of the reservoir pre-storage capacity is w j,k-1 ; Step 5.7) Let j=j+1, and repeat steps 5.2) to 5.6) until the upper limits of pre-storage capacity of all reservoirs are calculated, and the upper limits of pre-storage capacity of the cascade reservoir group are obtained.
2. A method for obtaining the upper limit of pre-storage water volume of a cascade reservoir group considering rainfall forecast information and its uncertainty according to claim 1, characterized in that: The step 1 is specifically as follows: Step 1.1) Classify the numerical rainfall forecasts, and then analyze the accuracy of the rainfall forecasts at different levels, different sub-basins, and different forecast periods, including accuracy, underreporting rate, and false alarm rate; Step 1.2) Based on the accuracy analysis of numerical rainfall forecasts in step 1.1), the forecast period and basin evolution time are compared to identify and select the available rainfall forecast information and forecast period.
3. A method for obtaining the upper limit of pre-storage water volume of a cascade reservoir group considering rainfall forecast information and its uncertainty according to claim 1, characterized in that: In step 5, the risk quantification formula and risk constraint based on rainfall forecast information are shown in formula (18) and formula (19): P j,k (Z d,j ,Z m,i )=P Z {Z[Z d,j ,X(p i )]≥Z m,i }; (19) P j,k (WITH d,j ,WITH m,i )≤P d,j ; (20) In the formula, Z[Z d,j ,X(p i )] is the jth reservoir with Z d,j To adjust the water level, adjust X(p i ) The high water level of flood caused by rainfall error; P d,j is the original design risk rate of the jth reservoir; P Z The rainfall frequency corresponding to the original designed flood control high water level is calculated in the flood control high water level-rainfall frequency curve.
