Cascade reservoir hydro-fluctuation period optimization scheduling scheme generation method and device

Abstract

The invention provides a method and a device for generating a cascade reservoir hydro-fluctuation period optimization scheduling scheme. The method comprises the following steps: collecting cascade reservoir engineering characteristic basic data and long-series natural runoff data; generalizing the maximum output flow constraint of the cascade reservoir in the hydro-fluctuation period under various influence factors by adopting a time sequence analysis method; constructing a cascade reservoir hydro-fluctuation period optimization scheduling model, and determining abandoned water quantity penalty factors of different typical years; solving the cascade reservoir hydro-fluctuation period optimization scheduling model to obtain an optimal water level process line of the cascade reservoir; studying and judging the size relation between the relative abandoned water quantity of the cascade reservoir hydro-fluctuation period and a preset threshold value one by one typical year, and adjusting an abandoned water quantity penalty factor according to the situation; and arranging reservoir water levels and overall to-be-eliminated reservoir capacities of different control time nodes of each cascade reservoir in different typical years to form a hydro-fluctuation period optimization scheduling scheme. The method has the advantages of being simple, practical, reasonable, reliable, easy to popularize and the like, and decision support can be provided for real-time scheduling of a cascade reservoir operation management department.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy and hydropower, and in particular to a method and device for generating an optimal dispatching scheme for a cascade reservoir during a drawdown period. Background Art

[0002] The drawdown period is characterized by a long duration, wide range of comprehensive utilization tasks and dispatching needs. Cascade reservoir power stations usually take maximum power generation and minimum water abandonment as the dispatching target of the drawdown period. For this reason, the reservoir water level is generally maintained at a high level in the early stage of drawdown to improve the water resource utilization efficiency of the power station; the drawdown is accelerated at the end of the drawdown to reduce the reservoir water level to the specified operating range during the flood season. Due to the long time span of the drawdown period, the power generation flow of the cascade reservoir power station will be subject to certain constraints due to multiple factors such as hydropower station maintenance, line maintenance, power grid peak regulation, water supply dispatching, and ecological dispatching; once the forecast of excessive water inflow is in the later stage, the cascade reservoir will have to open the gate to abandon water to reduce the reservoir water level to the specified operating range at the end of the drawdown period. At the same time, due to the similar hydrological rhythm and close hydraulic connection of cascade reservoirs, the drawdown process of the upstream cascade reservoir directly affects the water inflow level of the downstream cascade reservoir, and the high water level operation of the downstream cascade reservoir may affect the tail water level of the upstream cascade reservoir, thereby affecting the power generation head of the upstream power station.

[0003] At present, the optimization scheduling mode of cascade reservoir drawdown period is usually optimized with the goal of maximizing power generation or minimizing water abandonment, which fails to effectively coordinate the relationship between the two; and the optimization process does not effectively consider the impact of factors such as hydropower station maintenance, line maintenance, power grid peak regulation, water supply scheduling, and ecological scheduling on the maximum output and power generation flow of the power station, resulting in the optimization results not being very instructive for actual scheduling. In addition, due to the high dimension of the decision variables of the optimization target of cascade reservoirs, affected by the spatial coupling characteristics, the dimension of the decision variables in each time period is directly determined by the number of cascade hydropower stations, which leads to the "dimensionality curse" problem becoming a key factor restricting the practicality of the optimization algorithm. Summary of the invention

[0004] The purpose of the present invention is to provide a method and device for generating an optimized scheduling plan for a cascade reservoir during the drawdown period, which is used to solve the problems that the prior art fails to effectively coordinate the relationship between power generation and water abandonment, fails to consider influencing factors and has weak scheduling guidance and causes "dimensionality curse", and can provide technical support for guiding scheduling and operation decisions of cascade reservoirs during the drawdown period.

[0005] In order to achieve the above-mentioned object, in a first aspect, the present invention provides a method for generating an optimal scheduling scheme for a cascade reservoir drawdown period, comprising:

[0006] Step 1: Collect basic data on engineering characteristics of cascade reservoirs and long-series natural runoff data;

[0007] Step 2: Based on the basic data of engineering characteristics and long-term natural runoff data, the time series analysis method is used to generalize the maximum outflow constraints of cascade reservoirs under various influencing factors during the drawdown period;

[0008] Step 3: Construct an optimal scheduling model for the drawdown period of cascade reservoirs that takes into account both maximum power generation and minimum water abandonment, analyze the frequency of water inflow during the drawdown period, and determine the penalty factors for water abandonment in different typical years;

[0009] Step 4: Solve the optimal dispatching model of the cascade reservoir drawdown period by combining the stepwise optimization algorithm with the successive approximation algorithm to obtain the optimal water level process line of the cascade reservoir;

[0010] Step 5: According to the optimal water level process line and the penalty factors of abandoned water in different typical years, the relative abandoned water volume of the cascade reservoirs during the drawdown period is counted in each typical year. According to the relationship between the relative abandoned water volume and the preset threshold, the penalty factors of abandoned water volume are adjusted or the abandoned water volume and the storage capacity to be eliminated of the cascade reservoirs in each period are counted;

[0011] Step 6: According to the needs of scheduling and operation, determine the different control time nodes of each cascade reservoir in different typical years, calculate the reservoir water level and the overall storage capacity to be drawn at each control time node, and form an optimized scheduling plan for the drawdown period.

[0012] According to a method for generating an optimized scheduling scheme for a cascade reservoir drawdown period provided by the present invention, basic data of engineering characteristics include normal water storage level of each cascade reservoir, water level and storage capacity curve, start and end time of the drawdown period, full discharge flow, minimum discharge flow, reservoir water level variation constraint, tail water level relationship curve, and operating water level at the end of the drawdown period; multiple influencing factors include hydropower station maintenance, line maintenance, power grid peak regulation, water supply scheduling, and ecological scheduling.

[0013] According to a method for generating an optimal scheduling scheme for a cascade reservoir drawdown period provided by the present invention, step 2 specifically includes:

[0014] Determine the time step of the drawdown period of cascade reservoirs based on the time scale of long series natural runoff data;

[0015] According to the start and end time and time step of the decline period, the decline period is divided into n time periods, where n = TS / dt, TS = T1-T0; where T0 is the start time of the decline period, T1 is the end time of the decline period, TS is the duration of the decline period, and dt is the time step of the decline period;

[0016] If the operating years of the cascade reservoirs are greater than or equal to the preset value, the actual outflow flow data of the cascade reservoirs in each period of the drawdown period over the years are collected, and the maximum outflow flow constraint of the cascade reservoirs in each period is obtained as follows:

[0017]

[0018] In the formula, is the maximum outflow constraint of cascade reservoir i in period k, is the actual outflow of cascade reservoir i in the jth year in period k, and m is the number of years of operation of cascade reservoir i;

[0019] If the operating years of the cascade reservoirs are less than the preset value, then according to the dispatching requirements, the outflow constraints of the cascade reservoirs under various influencing factors in each period of the drawdown period over the years are analyzed, and the maximum outflow constraints of the cascade reservoirs in each period are obtained as follows:

[0020]

[0021] In the formula, The outflow constraint under the influence of hydropower station maintenance. is the outbound flow constraint under the influence of line maintenance, is the outbound flow constraint under the influence of power grid peak regulation, is the outflow constraint required by water supply scheduling, It is the outbound flow constraint required by ecological scheduling.

[0022] According to a method for generating an optimal scheduling scheme for a cascade reservoir drawdown period provided by the present invention, step 3 specifically includes:

[0023] The objective function of the optimal dispatching model for the drawdown period of cascade reservoirs is determined as follows:

[0024]

[0025] Where G is the overall target of optimal operation of cascade reservoirs, is the power generation output of cascade reservoir i in time period k, is the water discharge of cascade reservoir i in period k, is the penalty factor for the amount of water abandoned by cascade reservoir i in time period k, and R is the number of cascade reservoirs;

[0026] The constraints for determining the optimal dispatching model for the drawdown period of cascade reservoirs include:

[0027]

[0028] In the formula, is the water level of cascade reservoir i in time period k, is the water level of cascade reservoir i in period k+1, and are the minimum and maximum water level constraints of cascade reservoir i in period k, ΔZ i Indicates the maximum allowable reservoir water level fluctuation within adjacent time periods; is the outflow of cascade reservoir i in period k, and are the minimum and maximum outflow constraints of cascade reservoir i in time period k, respectively;

[0029] According to the long series of natural runoff data, frequency analysis is carried out based on the natural runoff in the drawdown period of each year. The years corresponding to different frequencies of natural runoff are selected to represent different typical years. Different penalty factors for abandoned water are set for different typical years.

[0030] According to a method for generating an optimal scheduling scheme for a cascade reservoir drawdown period provided by the present invention, step 4 specifically comprises:

[0031] Step 41, using an approximation algorithm to solve the initial water level process lines of the 1st to Rth cascade reservoirs;

[0032] Step 42, fix the initial water level process lines of the second to Rth cascade reservoirs, use a stepwise optimization algorithm to optimize the first cascade reservoir according to the overall goal of the cascade reservoir optimization scheduling, and obtain the optimal water level process line of the first cascade reservoir;

[0033] Step 43, fix the optimal water level process line after the first cascade reservoir is optimized and dispatched, and fix the initial water level process lines of the third to Rth cascade reservoirs, and continue to use the stepwise optimization algorithm to optimize the dispatch of the second cascade reservoir according to the overall goal of the cascade reservoir optimization dispatch, and obtain the optimal water level process line of the second cascade reservoir;

[0034] Step 44, and so on, traverse and optimize all cascade reservoirs to obtain the optimal water level process lines of all cascade reservoirs;

[0035] Step 45: When determining the optimal water level process lines of two adjacent cascade reservoirs, whether the values ​​of the total objective of the cascade reservoir optimization dispatch calculated twice adjacently meet the following conditions:

[0036]

[0037] δ≤δ lim (1)

[0038] In the formula, G l , G l+1 are the values ​​of the total objective of the optimal operation of the cascade reservoirs calculated twice in succession, δ lim is the preset optimization target discrimination threshold;

[0039] If the equation (1) is satisfied, the optimal water level process lines of the corresponding two adjacent cascade reservoirs are output; if the equation (1) is not satisfied, step 42 is executed until the equation (1) is satisfied.

[0040] According to a method for generating an optimal scheduling scheme for a cascade reservoir drawdown period provided by the present invention, step 5 specifically includes:

[0041] According to the optimal water level process line of cascade reservoirs, the relative water abandonment volume of cascade reservoirs during the drawdown period is counted in typical years to determine whether the relative water abandonment volume meets the following requirements:

[0042]

[0043] ΔW≤ΔW lim (2)

[0044] Where W 弃水 is the total amount of water abandoned during the drawdown period of the cascade reservoirs, W 来水 is the total water inflow of the downstream cascade reservoir during the drawdown period, ΔW lim is the preset threshold;

[0045] If formula (2) is satisfied, the drawdown water level and the storage capacity to be drawn during the drawdown period of the cascade reservoirs in the typical year are obtained; if formula (2) is not satisfied, the abandoned water volume and the storage capacity to be drawn in each period of the cascade reservoirs are counted, and for the period when the abandoned water volume is greater than zero, the penalty factor of the abandoned water volume of the cascade reservoirs in this period is increased, and step 4 is repeated until formula (2) is satisfied.

[0046] According to a method for generating an optimal scheduling scheme for a cascade reservoir drawdown period provided by the present invention, step 6 specifically includes:

[0047] Between the start time of the decline period and the end time of the decline period, according to the scheduling operation needs, select p control time nodes, and

[0048]

[0049] in, are p control time nodes;

[0050] The reservoir water level and overall storage capacity to be drawn are calculated at each control time node in each typical year to form an optimized scheduling plan for the drawdown period.

[0051] In a second aspect, the present invention further provides a device for generating an optimal scheduling scheme for a cascade reservoir drawdown period, comprising:

[0052] Collection unit, used to collect basic data on engineering characteristics of cascade reservoirs and long series of natural runoff data;

[0053] The generalization unit is used to generalize the maximum outflow constraints of cascade reservoirs under various influencing factors during the drawdown period using time series analysis method based on basic data of engineering characteristics and long series of natural runoff data;

[0054] Construct an analysis unit to build an optimal scheduling model for the drawdown period of cascade reservoirs that takes into account both maximum power generation and minimum water abandonment, conduct water inflow frequency analysis during the drawdown period, and determine the penalty factor for water abandonment in different typical years;

[0055] A solving unit is used to solve the optimal dispatching model of the drawdown period of the cascade reservoirs by combining a stepwise optimization algorithm with a successive approximation algorithm, and obtain the optimal water level process line of the cascade reservoirs;

[0056] A processing unit is used to count the relative amount of water abandoned during the drawdown period of the cascade reservoirs in each typical year according to the optimal water level process line and the penalty factors of water abandoned in different typical years, and to adjust the penalty factors of water abandoned or count the amount of water abandoned and the storage capacity to be eliminated in each period of the cascade reservoirs according to the relationship between the relative amount of water abandoned and a preset threshold value;

[0057] The generation unit is used to determine the different control time nodes of each cascade reservoir in different typical years according to the scheduling and operation needs, and to count the reservoir water level and the overall storage capacity to be eliminated obtained at each control time node to form an optimized scheduling plan for the drawdown period.

[0058] The present invention provides a method and device for generating an optimal scheduling scheme for a drawdown period of cascade reservoirs. The method comprises: collecting basic data on engineering characteristics of cascade reservoirs and long-series natural runoff data; generalizing the maximum outflow flow constraints of cascade reservoirs under various influencing factors during the drawdown period by using a time series analysis method; constructing an optimal scheduling model for a drawdown period of cascade reservoirs that takes into account both the maximum preference for power generation and the minimum preference for abandoned water, analyzing the frequency of water inflow during the drawdown period, and determining the penalty factors for abandoned water in different typical years; solving the optimal scheduling model for a drawdown period of cascade reservoirs by combining a stepwise optimization algorithm with a successive approximation algorithm to obtain an optimal water level process line for the cascade reservoirs; judging the relationship between the relative abandoned water volume and a preset threshold value during the drawdown period of the cascade reservoirs in each typical year according to the solved model results, and adjusting the penalty factors for abandoned water volume as appropriate; sorting out the reservoir water levels and the overall storage capacity to be abandoned at different control time nodes of each cascade reservoir in different typical years to form an optimal scheduling scheme for a drawdown period. The present invention has the advantages of being simple, practical, reasonable, reliable and easy to promote. According to the characteristics of the scheduling and application of the cascade reservoir drawdown period, under the premise of taking into account the dual goal optimization of increasing the power generation of the hydropower station and reducing the amount of abandoned water, the optimal drawdown process of the cascade reservoir under different water inflow conditions during the drawdown period is quickly given, thereby improving the power generation efficiency of the cascade reservoir, conforming to the operating characteristics of the actual scheduling of the cascade reservoir, and providing decision support for the real-time scheduling of the cascade reservoir operation management department. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] In the attached picture:

[0061] Figure 1 A flow chart of a method for generating an optimized scheduling scheme for cascade reservoir drawdown period according to the present invention;

[0062] Figure 2 It is a structural block diagram of a device for generating an optimized scheduling scheme for the drawdown period of cascade reservoirs according to the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] Some embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0065] The embodiment of the present invention provides a method for generating an optimized scheduling scheme for a cascade reservoir during the drawdown period, which can take into account the optimization goals of the power generation and water abandonment of the cascade reservoirs, reasonably describe the impact of multiple factors at different stages of the drawdown period of the cascade reservoirs on the power generation flow, reduce problems such as the dimensionality disaster of the cascade reservoir optimization, and meet the important scheduling needs of the scheduling and operation management of the cascade reservoir during the drawdown period.

[0066] See also Figure 1 The method for generating the optimal dispatching scheme of the cascade reservoir drawdown period includes the following steps:

[0067] Step 1: Collect basic data on engineering characteristics of cascade reservoirs and long-series natural runoff data;

[0068] Specifically, the basic data of engineering characteristics include the normal water storage level of each cascade reservoir, water level and storage capacity curve, start and end time of the drawdown period, full discharge, minimum discharge, reservoir water level fluctuation constraint, tail water level relationship curve, operating water level at the end of the drawdown period (i.e., the reservoir water level requirement that meets flood control safety at the end of the drawdown period), etc. The time scale of long series natural runoff data can be ten-day average or daily average.

[0069] Step 2: Based on the basic data of engineering characteristics and long-term natural runoff data, the time series analysis method is used to generalize the maximum outflow constraints of cascade reservoirs under various influencing factors during the drawdown period;

[0070] In some embodiments, the multiple influencing factors include: hydropower station maintenance, line maintenance, power grid peak regulation, water supply scheduling, ecological scheduling, etc.

[0071] Step 2 specifically includes:

[0072] Step 21, based on the time scale of the long series of natural runoff data, determine the time step dt of the drawdown period of the cascade reservoirs;

[0073] Step 22, according to the start and end time and time step of the decline period, divide the decline period into n time periods, where n=TS / dt, TS=T1-T0; where T0 is the start time of the decline period, T1 is the end time of the decline period, TS is the duration of the decline period, and dt is the time step of the decline period;

[0074] Step 23, analyzing the maximum outflow constraint in each period: for a cascade reservoir i with certain actual dispatching operation data (e.g., operation years m>=5), that is, if the operation years of the cascade reservoir are greater than or equal to the preset value, the actual outflow data of each period of the drawdown period of the cascade reservoirs are collected. j=1, 2, ..., m, k=1, 2, ..., n, and the maximum outflow constraint of cascade reservoir i in time period k is obtained as follows:

[0075]

[0076] In the formula, is the maximum outflow constraint of cascade reservoir i in period k, is the actual outflow of cascade reservoir i in the jth year in period k, and m is the number of years of operation of cascade reservoir i;

[0077] For cascade reservoirs whose operation years do not meet the above requirements, that is, if the operation years of the cascade reservoirs are less than the preset value, the outflow constraints under the influence of hydropower station maintenance at each period of the drawdown period of the cascade reservoirs are analyzed according to the scheduling needs. Outbound flow constraints under line maintenance Outbound flow constraints under the influence of power grid peak regulation Outflow constraints required by water supply scheduling Outbound flow constraints required by ecological scheduling The maximum outflow constraint of cascade reservoir i in period k is obtained as follows:

[0078]

[0079] Step 3: Construct an optimal scheduling model for the drawdown period of cascade reservoirs that takes into account both maximum power generation and minimum water abandonment, analyze the frequency of water inflow during the drawdown period, and determine the penalty factors for water abandonment in different typical years;

[0080] In some embodiments, step 3 specifically includes:

[0081] Step 31: Determine the objective function of the optimal dispatching model for the drawdown period of cascade reservoirs as:

[0082]

[0083] Where G is the overall target of cascade reservoir optimization operation, E 发电 is the power generation target for optimal dispatching of cascade reservoirs, W 弃水 is the target of water abandonment for optimal operation of cascade reservoirs, θ is the penalty factor for water abandonment, is the power generation output of cascade reservoir i in time period k, is the water discharge of cascade reservoir i in period k, is the penalty factor for water abandonment of cascade reservoir i in time period k, and R is the number of cascade reservoirs.

[0084] Step 32: Determine the constraints of the optimal scheduling model for the drawdown period of cascade reservoirs, including:

[0085] Reservoir water level constraints

[0086]

[0087] In the formula, is the water level of cascade reservoir i in time period k, is the water level of cascade reservoir i in period k+1, and are the minimum and maximum water level constraints of cascade reservoir i in period k, ΔZ i Indicates the maximum allowable reservoir water level fluctuation in adjacent time periods. Generally considered according to the dead water level; If there is no special requirement, the normal water storage level is generally considered at the initial moment.

[0088] Outbound flow constraints

[0089]

[0090] In the formula, is the outflow of cascade reservoir i in period k, and are the minimum and maximum outflow constraints of cascade reservoir i in time period k, respectively; According to the minimum downstream flow control approved by the dispatching regulations, The maximum outbound flow constraint control is determined in step 2.

[0091] Initial and final water level boundary conditions

[0092] The starting water level of cascade reservoir i in the initial period Usually controlled according to the normal water level; the controlled water level of cascade reservoir i at the end of drawdown period n It is usually considered based on the flood limit water level.

[0093] Step 33, based on the long series of natural runoff data, frequency analysis is performed based on the natural runoff volume in each drawdown period, and years corresponding to different frequencies of natural runoff volume are selected to represent different typical years, and different penalty factors for abandoned water are set for different typical years;

[0094] Optionally, for the long series of natural runoff data collected in step 1 (assuming that the data series has y years in total), based on the natural runoff in the drawdown period of each year, a frequency analysis is performed on the long series of runoff for y years, and years with 10%, 50%, and 90% frequency of natural runoff are selected to represent typical years of wet, normal, and dry seasons, respectively; different penalty factors for abandoned water are set for different typical years.

[0095] Step 4: Use a combination of stepwise optimization algorithm (POA) and successive approximation algorithm (DPSA) to solve the optimal scheduling model of the cascade reservoir drawdown period and obtain the optimal water level process line of the cascade reservoir;

[0096] Specifically, step 4 includes:

[0097] Step 41: Use the approximation algorithm (DP algorithm) to solve the initial water level process lines of the 1st to Rth cascade reservoirs.

[0098] Step 42, fix the initial water level process lines of the 2nd to Rth cascade reservoirs, use the POA algorithm to optimize the scheduling of the first cascade reservoir, the optimization target is the overall target of the cascade reservoir optimization scheduling determined in step 31, and obtain the optimal water level process line of the first cascade reservoir.

[0099] Step 43, fix the optimal water level process line of the first cascade reservoir after optimized scheduling, and fix the initial water level process lines of the third to Rth cascade reservoirs, and continue to use the POA algorithm to optimize the scheduling of the second cascade reservoir according to the overall goal of optimized scheduling of cascade reservoirs, and obtain the optimal water level process line of the second cascade reservoir.

[0100] Step 44, and so on, traverse and optimize all cascade reservoirs to obtain the optimal water level process lines of all cascade reservoirs.

[0101] Step 45: When determining the optimal water level process lines of two adjacent cascade reservoirs, whether the values ​​of the total objective of the cascade reservoir optimization dispatch calculated twice adjacently meet the following conditions:

[0102]

[0103] δ≤δ lim (1)

[0104] In the formula, G l , G l+1are the values ​​of the total objective of the optimal operation of the cascade reservoirs calculated twice in succession, δ lim is the preset optimization target discrimination threshold;

[0105] If the formula (1) is satisfied, the optimal water level process lines of the corresponding two adjacent cascade reservoirs are output; if the formula (1) is not satisfied, step 42 is executed until the formula (1) is satisfied.

[0106] The present invention introduces a preset optimization target discrimination threshold δ lim , to determine the value of the total target of cascade reservoir optimization operation G calculated this time l+1 The value G of the total target of cascade reservoir optimization operation obtained in the previous calculation l Compared with whether the accuracy requirements are met, if δ≤δ lim Then stop the calculation and output the optimal water level process line of the cascade reservoir, that is, the state change process; otherwise return to step 42 to continue iterative calculation until the iteration requirements are met.

[0107] Step 5: According to the optimal water level process line and the penalty factors of abandoned water in different typical years, the relative abandoned water volume of the cascade reservoirs during the drawdown period is counted in each typical year. According to the relationship between the relative abandoned water volume and the preset threshold, the penalty factors of abandoned water volume are adjusted or the abandoned water volume and the storage capacity to be eliminated of the cascade reservoirs in each period are counted;

[0108] In some embodiments, step 5 specifically includes:

[0109] Step 51: According to the optimal water level process line of the cascade reservoirs, the relative water abandonment volume of the cascade reservoirs during the drawdown period is counted in each typical year to determine whether the relative water abandonment volume meets the following requirements:

[0110]

[0111] ΔW≤ΔW lim (2)

[0112] Where W 弃水 is the total amount of water abandoned during the drawdown period of the cascade reservoirs, W 来水 is the total water inflow of the downstream cascade reservoir during the drawdown period, ΔW lim is the preset threshold;

[0113] Step 52, if formula (2) is satisfied, the drawdown water level and the storage capacity to be eliminated of the cascade reservoirs during the drawdown period of the typical year are obtained; if formula (2) is not satisfied, the abandoned water volume and the storage capacity to be eliminated of the cascade reservoirs in each time period are counted, and for the time period when the abandoned water volume is greater than zero, the abandoned water volume penalty factor of the cascade reservoirs in the time period is increased, and step 4 is repeated until formula (2) is satisfied.

[0114] The present invention introduces an acceptable threshold value ΔW for the relative amount of water abandoned during the drawdown period of cascade reservoirs lim, that is, the preset threshold value. Based on step 4, the amount of water abandoned during the drawdown period of the cascade reservoirs is counted in each typical year. 弃水 Determine whether the relative water discharge volume ΔW of the cascade reservoirs in a typical year is less than or equal to ΔW lim , if ΔW≤ΔW lim , then the simulation results of the cascade reservoirs in this typical year meet the final requirements; if ΔW>ΔW lim , further optimize the cascade decline process of this typical year, that is, for ΔW>ΔW lim In a typical year, the amount of water abandoned by cascade reservoirs in statistical period k (k = 1, 2, ..., n) is δW 弃水,l and the storage capacity to be eliminated of the cascade reservoirs ΔV l ; For δW 弃水,l >0, increase the penalty factor for the amount of water abandoned by the cascade reservoir in period k (It can be considered as each increase of dθ), and repeat step 4 to perform the optimal scheduling simulation calculation for the typical year until ΔW ≤ ΔW lim .

[0115] Step 6: According to the needs of scheduling and operation, determine the different control time nodes of each cascade reservoir in different typical years, calculate the reservoir water level and the overall storage capacity to be drawn at each control time node, and form an optimized scheduling plan for the drawdown period.

[0116] In some embodiments, step 6 specifically includes:

[0117] Step 61: between the start time T0 of the decline period and the end time T1 of the decline period, according to the scheduling operation requirements, select p important control time nodes and

[0118]

[0119] Step 62: For each of the selected typical years of high water, normal water and low water, the reservoir water level and the overall storage capacity to be drawn at each control time node of each typical year are statistically analyzed to form an optimized scheduling plan for the drawdown period.

[0120] Based on the same inventive concept, another embodiment of the present invention provides a device for generating an optimal scheduling scheme for a cascade reservoir drawdown period, which corresponds to the method of the aforementioned embodiment, such as Figure 2 As shown, the device comprises:

[0121] Collection unit, used to collect basic data on engineering characteristics of cascade reservoirs and long series of natural runoff data;

[0122] The generalization unit is used to generalize the maximum outflow constraints of cascade reservoirs under various influencing factors during the drawdown period using time series analysis method based on basic data of engineering characteristics and long series of natural runoff data;

[0123] Construct an analysis unit to build an optimal scheduling model for the drawdown period of cascade reservoirs that takes into account both maximum power generation and minimum water abandonment, conduct water inflow frequency analysis during the drawdown period, and determine the penalty factor for water abandonment in different typical years;

[0124] A solving unit is used to solve the optimal dispatching model of the drawdown period of the cascade reservoirs by combining a stepwise optimization algorithm with a successive approximation algorithm, and obtain the optimal water level process line of the cascade reservoirs;

[0125] A processing unit is used to count the relative amount of water abandoned during the drawdown period of the cascade reservoirs in each typical year according to the optimal water level process line and the penalty factors of water abandoned in different typical years, and to adjust the penalty factors of water abandoned or count the amount of water abandoned and the storage capacity to be eliminated in each period of the cascade reservoirs according to the relationship between the relative amount of water abandoned and a preset threshold value;

[0126] The generation unit is used to determine the different control time nodes of each cascade reservoir in different typical years according to the scheduling and operation needs, and to count the reservoir water level and the overall storage capacity to be eliminated obtained at each control time node to form an optimized scheduling plan for the drawdown period.

[0127] The following is a specific embodiment of the present invention.

[0128] Taking four cascade reservoirs in the Jinsha River Basin as an example, the method for generating an optimal scheduling scheme for the drawdown period of cascade reservoirs provided by the present invention is applied, and the specific process is as follows:

[0129] 1. Collect the normal water storage level, water level and storage capacity curve, start and end time of the drawdown period, full discharge, minimum discharge, reservoir water level fluctuation constraint, tail water level relationship curve, operating water level at the end of the drawdown period and long series of natural runoff data (1959-2014) of the four cascade reservoirs. Among them, some engineering characteristic parameters such as the normal water storage level, start and end time of the drawdown period, full discharge, minimum discharge, and operating water level at the end of the drawdown period of the four cascade reservoirs are shown in Table 1.

[0130] Table 1. Some engineering characteristic parameters of four cascade reservoirs

[0131]

[0132] 2. Determine the calculation time step dt as ten days, and the calculation period is n=21; by collecting the historical dispatching and operation data of cascade reservoirs and the impact of different dispatching demands and maintenance on the outflow, analyze and generalize the maximum outflow constraints of different periods and cascade reservoirs under the influence of multiple factors such as hydropower station maintenance, line maintenance, power grid peak regulation, water supply dispatching, and ecological dispatching during the drawdown period, as shown in Table 2 for details.

[0133] Table 2. Maximum power generation flow constraint coefficients of four cascade reservoirs

[0134]

[0135]

[0136] 3. Construct an optimal scheduling model for the drawdown period of cascade reservoirs that takes into account the maximum preference for power generation and the minimum preference for water abandonment; determine the constraints such as reservoir water level, outflow flow, and initial and final water levels during the drawdown period of cascade reservoirs; conduct frequency analysis according to the total runoff during the drawdown period in a long series of natural runoff for 56 years from 1959 to 2014, and select 10%, 50%, and 90% frequency years of natural runoff in 1978, 1961, and 1982 to represent typical years of high, medium, and low seasons, respectively; preliminarily determine the penalty factor θ for water abandonment in different periods of each cascade reservoir in different typical years: from December 1 to April 30, that is, θ = 0.02 for periods 1 to 15; from May 1 to June 30, that is, θ = 0.08 for calculation periods 16 to 21; the above is referred to as "Scheme 1".

[0137] 4. Determine the optimization target discrimination threshold δ lim =0.01%. The constructed optimal dispatching model for the drawdown period of cascade reservoirs is used to simulate different typical years, and the simulation results of different typical years are output and sorted. Table 3 shows the power generation and water abandonment of each cascade reservoir during the drawdown period in the typical year of Scheme 1.

[0138] Table 3. Power generation and water abandonment of each cascade reservoir during the drawdown period in a typical year of Scheme 1

[0139]

[0140] 5. Analyze the amount of water abandoned in each typical year. For 1982, there was no water abandoned during the drawdown period, so Scheme 1 is the recommended scheme. For the typical years 1978 and 1961 in which water was abandoned during the drawdown period in the simulation results of Scheme 1, determine the acceptable threshold ΔW for the relative amount of water abandoned during the drawdown period of the cascade reservoirs. lim =2%, the relative amount of water abandonment ΔW in 1978 and 1961 is calculated to be 6.53% and 3.32% respectively.

[0141] Further increase the penalty factor θ of abandoned water (each time increase dθ, where the single step dθ of calculation period 1 to 15 is 0.02, and the single step dθ of calculation period 16 to 21 is 0.08). After iterative calculation and judgment, when θ of calculation period 1 to 15 is 0.06 and θ of calculation period 16 to 21 is 0.24 (referred to as "Scheme 2"), the relative abandoned water volume ΔW of cascade reservoirs during the drawdown period in 1961 is 0.01%, which is reduced to the acceptable threshold range. Scheme 2 is its recommended scheme; the relative abandoned water volume ΔW of cascade reservoirs during the drawdown period in 1978 is 4.36%, which requires further iterative calculation. Table 4 shows the total power generation and abandoned water volume of each cascade reservoir during the drawdown period in a typical year of Scheme 2. Comparing Table 3 and Table 4, for 1978, the power generation decreased by 0.2%, but the abandoned water volume decreased by 33.2%; for 1961, the power generation decreased by 0.2%, and the abandoned water volume decreased by 99.6%.

[0142] Table 4. Power generation and water abandonment of each cascade reservoir during the drawdown period in a typical year of Scheme 2

[0143]

[0144] For the flood year of 1978, the penalty factor θ of abandoned water is further increased. After iterative calculation and judgment, when θ=0.8 for calculation period 1 to 15 and θ=3.2 for calculation period 16 to 21 (referred to as "Scheme 3"), the relative abandoned water volume ΔW of the cascade reservoir drawdown period in 1978 is 2.00%, which is within the acceptable threshold range. Scheme 3 is the recommended scheme. Table 5 shows the total power generation and abandoned water volume of each cascade reservoir during the drawdown period in a typical year of Scheme 3. Comparing Table 3 and Table 5, for 1978, the power generation decreased by 1.7%, but the abandoned water volume decreased by 69.2%.

[0145] Table 5. Power generation and water abandonment during drawdown period of each cascade reservoir in a typical year of Scheme 3

[0146]

[0147]

[0148] 6. Select important control time nodes such as January 1, February 1, March 1, April 1, May 1, June 1, June 10, and June 20; sequentially count the reservoir water level, overall storage capacity to be eliminated, and other indicators calculated at the above different time nodes under the recommended schemes for each typical year to form a recommended optimized scheduling scheme for the drawdown period. See Table 6 for details.

[0149] Table 6. Distribution of water levels and overall storage capacity to be eliminated for each cascade reservoir in the recommended scheme

[0150]

[0151]

[0152] Note: The negative value of the overall storage capacity to be eliminated in the cascade indicates the storage capacity value below the flood limit water level.

[0153] In summary, the method and device for generating an optimal scheduling scheme for the drawdown period of cascade reservoirs provided by the present invention have the following beneficial effects:

[0154] 1. Compared with a single optimization scheduling method that maximizes power generation or minimizes water abandonment, the optimization scheduling method for coordinating power generation and water abandonment proposed in the present invention can better take into account both types of optimization objectives, and can also adjust the optimization scheduling plan according to the amount of water abandonment.

[0155] 2. It can fully combine the historical dispatching and operation characteristics with the actual dispatching needs, and summarize and provide the maximum outflow constraints of different periods and different cascade reservoirs under the influence of multiple factors such as hydropower station maintenance, line maintenance, power grid peak regulation, water supply dispatching, and ecological dispatching during the drawdown period, which conforms to the actual dispatching and application characteristics of cascade reservoirs.

[0156] 3. The present invention solves the optimal scheduling model of the drawdown period of cascade reservoirs by combining a step-by-step optimization algorithm (POA) with a successive approximation algorithm (DPSA). While eliminating the problem of a long calculation cycle caused by the "curse of dimensionality", the optimization result can be made as close to the global optimal solution as possible. The method has the advantages of good solution effect and fast calculation speed.

[0157] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the embodiments disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for generating an optimal scheduling scheme for a cascade reservoir drawdown period, characterized in that: include: Step 1: Collect basic data on engineering characteristics of cascade reservoirs and long-series natural runoff data; Step 2: Based on the basic data of engineering characteristics and long-series natural runoff data, a time series analysis method is used to generalize the maximum outflow constraints of the cascade reservoirs under various influencing factors during the drawdown period; Step 3: Construct an optimal scheduling model for the drawdown period of cascade reservoirs that takes into account both maximum power generation and minimum water abandonment, analyze the frequency of water inflow during the drawdown period, and determine the penalty factors for water abandonment in different typical years; Step 4: Solve the drawdown optimization scheduling model of the cascade reservoirs by combining a stepwise optimization algorithm with a successive approximation algorithm to obtain the optimal water level process line of the cascade reservoirs; Step 5: According to the optimal water level process line and the penalty factors of abandoned water in different typical years, the relative abandoned water volume of the cascade reservoirs during the drawdown period is counted in each typical year; according to the relationship between the relative abandoned water volume and the preset threshold, the abandoned water volume penalty factor is adjusted or the abandoned water volume and the storage capacity to be eliminated of the cascade reservoirs in each period are counted; Step 6: According to the needs of scheduling and operation, determine the different control time nodes of each cascade reservoir in different typical years, calculate the reservoir water level and the overall storage capacity to be drawn at each control time node, and form an optimized scheduling plan for the drawdown period.

2. The method for generating an optimal scheduling scheme for the drawdown period of cascade reservoirs according to claim 1 is characterized in that: The basic data of engineering characteristics include the normal water storage level of each cascade reservoir, water level and storage capacity curve, the start and end time of the drawdown period, full discharge flow, minimum discharge flow, reservoir water level variation constraint, tail water level relationship curve, and operating water level at the end of the drawdown period; the various influencing factors include hydropower station maintenance, line maintenance, power grid peak regulation, water supply scheduling, and ecological scheduling.

3. The method for generating an optimal scheduling scheme for the drawdown period of cascade reservoirs according to claim 2 is characterized in that: The step 2 specifically includes: Determine the time step of the drawdown period of cascade reservoirs based on the time scale of long series natural runoff data; According to the start and end time and time step of the decline period, the decline period is divided into n time periods, where n = TS / dt, TS = T1-T0; where T0 is the start time of the decline period, T1 is the end time of the decline period, TS is the duration of the decline period, and dt is the time step of the decline period; If the operating years of the cascade reservoirs are greater than or equal to the preset value, the actual outflow flow data of the cascade reservoirs in each period of the drawdown period over the years are collected, and the maximum outflow flow constraint of the cascade reservoirs in each period is obtained as follows: In the formula, is the maximum outflow constraint of cascade reservoir i in period k, is the actual outflow of cascade reservoir i in the jth year in period k, and m is the number of years of operation of cascade reservoir i; If the operating years of the cascade reservoirs are less than the preset value, then according to the dispatching requirements, the outflow constraints of the cascade reservoirs under various influencing factors in each period of the drawdown period over the years are analyzed, and the maximum outflow constraints of the cascade reservoirs in each period are obtained as follows: In the formula, The outflow constraint under the influence of hydropower station maintenance. is the outbound flow constraint under the influence of line maintenance, is the outbound flow constraint under the influence of power grid peak regulation, is the outflow constraint required by water supply scheduling, It is the outbound flow constraint required by ecological scheduling.

4. The method for generating an optimal scheduling scheme for the drawdown period of cascade reservoirs according to claim 3 is characterized in that: The step 3 specifically includes: The objective function of the optimal dispatching model for the cascade reservoir drawdown period is determined as: Where G is the overall target of optimal operation of cascade reservoirs, is the power generation output of cascade reservoir i in time period k, is the water discharge of cascade reservoir i in period k, is the penalty factor for the amount of water abandoned by cascade reservoir i in time period k, and R is the number of cascade reservoirs; The constraints for determining the optimal dispatching model of the cascade reservoir drawdown period include: In the formula, is the water level of cascade reservoir i in time period k, is the water level of cascade reservoir i in period k+1, and are the minimum and maximum water level constraints of cascade reservoir i in period k, ΔZ i Indicates the maximum allowable reservoir water level fluctuation within adjacent time periods; is the outflow of cascade reservoir i in period k, and are the minimum and maximum outflow constraints of cascade reservoir i in time period k, respectively; According to the long series of natural runoff data, frequency analysis is carried out based on the natural runoff in the drawdown period of each year. Years corresponding to different frequencies of natural runoff are selected to represent different typical years. Different penalty factors for abandoned water are set for different typical years.

5. The method for generating an optimal scheduling scheme for the drawdown period of cascade reservoirs according to claim 4 is characterized in that: The step 4 specifically includes: Step 41, using an approximation algorithm to solve the initial water level process lines of the 1st to Rth cascade reservoirs; Step 42, fixing the initial water level process lines of the second to Rth cascade reservoirs, using a stepwise optimization algorithm to optimize the first cascade reservoir according to the overall goal of the cascade reservoir optimization scheduling, and obtaining the optimal water level process line of the first cascade reservoir; Step 43, fix the optimal water level process line after the first cascade reservoir is optimized and dispatched, and fix the initial water level process lines of the third to Rth cascade reservoirs, and continue to use the stepwise optimization algorithm to optimize the dispatch of the second cascade reservoir according to the overall goal of the cascade reservoir optimization dispatch, and obtain the optimal water level process line of the second cascade reservoir; Step 44, and so on, traverse and optimize all cascade reservoirs to obtain the optimal water level process lines of all cascade reservoirs; Step 45: When determining the optimal water level process lines of two adjacent cascade reservoirs, whether the values ​​of the total objective of the cascade reservoir optimization dispatch calculated twice adjacently satisfy the following conditions: δ≤δ lim (1) In the formula, G l , G l+1 are the values ​​of the total target of the optimal dispatching of the cascade reservoirs calculated twice in succession, δ lim It is the preset optimization target discrimination threshold; If the formula (1) is satisfied, the optimal water level process lines of the corresponding two adjacent cascade reservoirs are output; if the formula (1) is not satisfied, step 42 is executed until the formula (1) is satisfied.

6. The method for generating an optimal scheduling scheme for the drawdown period of cascade reservoirs according to claim 5 is characterized in that: The step 5 specifically includes: According to the optimal water level process line of cascade reservoirs, the relative water abandonment volume of cascade reservoirs during the drawdown period is counted in typical years to determine whether the relative water abandonment volume meets the following requirements: ΔW≤ΔW lim (2) Where W 弃水 is the total amount of water abandoned during the drawdown period of the cascade reservoirs, W 来水 is the total water inflow of the downstream cascade reservoir during the drawdown period, ΔW lim is the preset threshold; If formula (2) is satisfied, the drawdown water level and the storage capacity to be drawn during the drawdown period of the cascade reservoirs in the typical year are obtained; if formula (2) is not satisfied, the abandoned water volume and the storage capacity to be drawn in each period of the cascade reservoirs are counted, and for the period when the abandoned water volume is greater than zero, the penalty factor of the abandoned water volume of the cascade reservoirs in this period is increased, and step 4 is repeated until formula (2) is satisfied.

7. The method for generating an optimal scheduling scheme for the drawdown period of cascade reservoirs according to claim 6 is characterized in that: The step 6 specifically includes: Between the start time of the decline period and the end time of the decline period, according to the scheduling operation needs, select p control time nodes, and in, are p control time nodes; The reservoir water level and overall storage capacity to be drawn are calculated at each control time node in each typical year to form an optimized scheduling plan for the drawdown period.

8. A device for generating an optimal scheduling scheme for a cascade reservoir drawdown period, characterized in that: include: Collection unit, used to collect basic data on engineering characteristics of cascade reservoirs and long series of natural runoff data; A generalization unit is used to generalize the maximum outflow constraints of the cascade reservoirs under various influencing factors during the drawdown period by using a time series analysis method based on the basic data of the engineering characteristics and a long series of natural runoff data; Construct an analysis unit to build an optimal scheduling model for the drawdown period of cascade reservoirs that takes into account both maximum power generation and minimum water abandonment, conduct water inflow frequency analysis during the drawdown period, and determine the penalty factor for water abandonment in different typical years; A solving unit, used for solving the drawdown optimization scheduling model of the cascade reservoirs by combining a stepwise optimization algorithm with a successive approximation algorithm, and obtaining an optimal water level process line of the cascade reservoirs; A processing unit is used to count the relative amount of water abandoned during the drawdown period of the cascade reservoirs in each typical year according to the optimal water level process line and the penalty factors for abandoned water in different typical years, and to adjust the penalty factors for abandoned water or count the amount of water abandoned and the storage capacity to be eliminated in each period of the cascade reservoirs according to the relationship between the relative amount of water abandoned and a preset threshold value; The generation unit is used to determine the different control time nodes of each cascade reservoir in different typical years according to the scheduling and operation needs, and to count the reservoir water level and the overall storage capacity to be eliminated obtained at each control time node to form an optimized scheduling plan for the drawdown period.

Images