Hydropower station group optimization scheduling dimension reduction method based on feasible region search mapping

Abstract

The invention relates to a hydropower station group optimization scheduling dimensionality reduction method based on feasible region search mapping, comprising the following steps: (1) sorting and classifying scheduling operation constraints of hydropower stations in different periods, and dividing constraint conditions related to hydropower station group operation into three levels; (2) determining decision variables of hydropower station scheduling in different periods and under different conditions, namely a mapping relation between a feasible region of ex-warehouse flow and known operation conditions and scheduling constraints; and (3) when the algorithm optimizes the ex-warehouse flow of the hydropower station in a certain time period, querying a corresponding decision variable feasible region in the database according to the time period, the in-warehouse flow and the initial storage capacity of the time period, and only carrying out optimal solution search in the feasible region range. According to the method, the constraint condition sets of the hydropower station group in different periods are converted into the feasible regions of the decision variables in advance, so that the search space of the optimization algorithm can be effectively reduced, and the calculation efficiency of the optimization algorithm is greatly improved.

Description

Technical Field

[0001] The invention relates to a dimensionality reduction method for optimizing the dispatching of a hydropower station group based on feasible domain search mapping, and belongs to the technical field of reservoir group dispatching. Background Art

[0002] With the continuous advancement of hydropower construction, large hydropower stations such as Wudongde and Baihetan have been completed and put into operation one after another. The scale of hydropower stations that need to be managed uniformly is also expanding, and the difficulty of dispatching management is growing exponentially. At present, the optimization dispatching model of hydropower station groups is generally constructed, and dynamic programming, intelligent algorithms and other methods are used to solve the dispatching scheme that maximizes the operating benefits of the hydropower station group or minimizes the dispatching risk, so as to improve the dispatching management level and reduce human bias in the decision-making process.

[0003] The optimal dispatching of hydropower stations is essentially a complex decision-making optimization problem under the coupling of high-dimensional multiple constraints. It has always been recognized as one of the most challenging optimization problems at home and abroad. The algorithm usually has the problem of "curse of dimensionality" in the calculation process. In order to meet the timeliness and accuracy requirements of the preparation of the optimal dispatching plan, it is necessary to design a scientific and reasonable dimensionality reduction method to effectively deal with the efficiency problem of solving large-scale hydropower system models. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a dimensionality reduction method for optimizing the scheduling of a group of hydropower stations based on feasible domain search mapping. By converting the constraint condition sets of the hydropower station group at different periods into the feasible domain of decision variables in advance, the search space of the optimization algorithm can be effectively reduced, and the computational efficiency of the optimization algorithm can be greatly improved.

[0005] The technical solution of the present invention is as follows:

[0006] A dimensionality reduction method for optimal scheduling of hydropower station groups based on feasible domain search mapping, the steps are as follows:

[0007] (1) The dispatching and operating constraints of hydropower stations in different periods are sorted and classified, and based on the manual experience of dispatchers, the constraints involved in the operation of hydropower stations are divided into three levels;

[0008] (2) Determine the decision variables for hydropower station dispatching under different periods and conditions, that is, the mapping relationship between the feasible domain of outflow flow and known operating conditions and dispatching constraints, and store them in the database;

[0009] (3) When optimizing the outflow of a hydropower station during a certain period of time, the algorithm (dynamic programming algorithm, intelligent algorithm, etc.) can query the feasible domain of the corresponding decision variables in the database according to the period of time, the inflow and the initial storage capacity of the period, and search for the optimal solution only within the feasible domain, so as to significantly improve the computational efficiency of the algorithm.

[0010] According to the preferred embodiment of the present invention, in step (1), the first-level constraint is the constraint that needs to be met to ensure the safe and stable operation of the reservoir, and must be followed, including the water level constraint and storage capacity constraint specified in the reservoir regulations; the second-level constraint is the constraint on the outflow of the reservoir by the flood control, ecology, water supply, shipping and other departments to ensure the downstream flood control safety and improve the ecological environment of the river basin; the third-level constraint is the requirement of the power generation department for the output of the hydropower station to meet the needs of peak load and frequency regulation of the power grid. According to the actual situation of the hydropower station dispatch, when there is a contradiction in the constraints in the optimization dispatch model, the higher-level constraints are preferentially satisfied.

[0011] According to a preferred embodiment of the present invention, in step (2), for a time period t, the inflow flow I of the hydropower station in the time period is known. t and initial storage capacity V t , the outflow of the hydropower station in the previous period Q t-1 and output N t-1 , recorded as the known operating condition L0;

[0012] The set of constraints in time period t is collectively recorded as scheduling constraint L t , the decision variables of the hydropower station dispatch in period t (discharge flow Q t )’s feasible region W and L0, L t The mapping relationship between them is recorded as W = f(L0,L t ). f involves multiple nonlinear reservoir characteristic curves and logical relationship judgments. The specific form of its analytical expression is difficult to give. Here, the discrete value method is used to obtain the multidimensional space surface corresponding to f. The calculation process is as follows:

[0013] ① Given the inflow and initial storage capacity of a hydropower station in a certain period, the constraints of different levels in the period are converted into constraints for the outflow in the period;

[0014] ② Couple the constraints obtained in step ① to obtain the feasible domain of the outbound flow during this period;

[0015] ③ Repeat steps ① and ② to calculate and store the feasible domain of outbound flow under different time periods, different inbound flow rates and initial storage capacity of the time period.

[0016] According to the preferred embodiment of the present invention, in step ①, specifically,

[0017] For time period t, based on water balance, the first-level constraints (water level and reservoir capacity constraints) are converted into the outflow flow Q of the hydropower station in time period t: t Constraints U1:

[0018]

[0019]

[0020]

[0021] In the formula, Q t represents the average outflow of the hydropower station during period t; They represent the lower and upper limits of the outbound flow corresponding to the first-level constraint; V t+1,min 、V t+1,max They represent the lower and upper limits of storage capacity at the end of time period t respectively; Δt represents the length of the time period.

[0022] According to the preferred embodiment of the present invention, in step ①, for time period t, the secondary constraint is used as the outflow flow Q of the hydropower station during time period t. t Constraints U2:

[0023]

[0024] In the formula, They represent the lower and upper limits of the outbound flow corresponding to the secondary constraints respectively; Q t-1 represents the average outflow of the hydropower station during the period t-1; ΔQ t,re , ΔQ t,add They represent the maximum increase and decrease of outbound flow in period t respectively.

[0025] According to the preferred embodiment of the present invention, in step ①, for time period t, the third-level constraint (output constraint) is converted into the outflow flow Q of the hydropower station in time period t. t Constraints U3:

[0026] U3=U 3,1 ∪U 3,2

[0027]

[0028]

[0029]

[0030]

[0031] Where U 3,1 , U 3,2 Represents two subsets of U3; N t,min 、N t,max They represent the lower and upper limits of the hydropower station output during period t; f NQ (·) represents the output function relationship of the hydropower station; They represent the two outflow flow values ​​corresponding to the lower limit of the hydropower station output (the hydropower station output function relationship f NQ (·) Usually parabolic, 1 output value corresponds to 2 outflows); They respectively represent the two outflow flow values ​​corresponding to the upper limit of the hydropower station output.

[0032] According to the preferred embodiment of the present invention, in step ②, specifically, the constraints of different levels for different variables are converted into the outbound flow Q in period t t The constraints {U1,U2,U3}, let the set W1=U1,W2=U1∩U2,W3=U1∩U2∩U3;

[0033] Initial storage capacity V t Can be changed within a given range, traversing V t The possible values ​​of {W1, W2, W3} are calculated. If all W3 are empty sets, it is considered that the three-level constraint cannot be satisfied. Otherwise, the outbound flow Q in period t is t The feasible domain W=f(L0,L t )=W3; if all W2 and W3 are empty sets, it is considered that the secondary constraint cannot be satisfied, and the feasible domain W=f(L0,L t )=W1, otherwise the feasible domain W=f(L0,L t )=W2.

[0034] According to the preferred embodiment of the present invention, in step ③, specifically:

[0035] Let time period t = 1, given the scheduling constraint L t , inbound flow I t Discretize within the possible range, the number of discrete points is N, and the initial storage capacity of the time period is V t Discretize within the possible range, the number of discrete points is M; according to steps ① and ②, calculate the different I in time period t t and V t Combined outbound flow Q t The feasible domain W is stored in the database; let t = t + 1, and repeat the above steps until t is greater than the number of scheduling period time periods T.

[0036] The present invention realizes the transformation and rapid query of constraint condition set into feasible domain of decision variables, which can effectively reduce the search space of optimization algorithm, improve the calculation efficiency of algorithm, and provide a new dimensionality reduction method for solving the optimization scheduling problem of large-scale hydropower station group.

[0037] The beneficial effects of the present invention are:

[0038] The present invention can effectively reduce the search space of the optimization algorithm and greatly improve the calculation efficiency of the optimization algorithm by converting the constraint condition sets of the hydropower station group at different periods into the feasible domain of decision variables in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 The figure is a calculation flow chart of the feasible domain search mapping of the present invention. DETAILED DESCRIPTION

[0040] The present invention will be further described below by way of embodiments in conjunction with the accompanying drawings, but is not limited thereto.

[0041] Embodiment 1:

[0042] This embodiment provides a method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping, and the steps are as follows:

[0043] (1) The constraints of the hydropower station at different times are sorted and classified, and based on the manual experience of the dispatchers, the constraints involved in the operation of the hydropower station group are divided into three levels; the first-level constraints are the constraints that must be met to ensure the safe and stable operation of the reservoir, which must be followed, including the water level constraints and storage capacity constraints stipulated in the reservoir regulations; the second-level constraints are the constraints on the outflow of the reservoir by the flood control, ecology, water supply, shipping and other departments to ensure the safety of downstream flood control and improve the ecological environment of the river basin; the third-level constraints are the requirements of the power generation department for the output of the hydropower station to meet the needs of power grid peak regulation and frequency regulation. According to the actual situation of the hydropower station dispatch, when there is a contradiction in the constraints in the optimization dispatch model, the higher-level constraints are given priority.

[0044] (2) Determine the mapping relationship between the feasible domain of the decision variables (outflow) of the hydropower station dispatch under different periods and conditions and the known operating conditions and dispatch constraints, and store them in the database;

[0045] For time period t, the inflow flow I of the hydropower station in this period is known t and initial storage capacity V t , the outflow of the hydropower station in the previous period Q t-1 and output N t-1 , recorded as the known operating condition L0;

[0046] The set of constraints in time period t is collectively recorded as scheduling constraint L t , the decision variables of the hydropower station dispatch in period t (discharge flow Q t ) The feasible region W and L0, L t The mapping relationship between them is recorded as W = f(L0,L t ). f involves multiple nonlinear reservoir characteristic curves and logical relationship judgments. The specific form of its analytical expression is difficult to give. Here, the discrete value method is used to obtain the multidimensional space surface corresponding to f. The calculation process is as follows:

[0047] ① Given the inflow flow rate and initial storage capacity of a hydropower station in a certain period, the constraints of different levels in the period are converted into constraints U1, U2 and U3 for the outflow flow in the period;

[0048] For time period t, based on water balance, the first-level constraints (water level and reservoir capacity constraints) are converted into the outflow flow Q of the hydropower station in time period t:t Constraints U1:

[0049]

[0050]

[0051]

[0052] In the formula, Q t represents the average outflow of the hydropower station during period t; They represent the lower and upper limits of the outbound flow corresponding to the first-level constraint; V t+1,min 、V t+1,max They represent the lower and upper limits of storage capacity at the end of time period t respectively; Δt represents the length of the time period.

[0053] For time period t, the secondary constraint is taken as the outflow flow Q of the hydropower station during time period t t Constraints U2:

[0054]

[0055] In the formula, They represent the lower and upper limits of the outbound flow corresponding to the secondary constraints respectively; Q t-1 represents the average outflow of the hydropower station during the period t-1; ΔQ t,re , ΔQ t,add They represent the maximum increase and decrease of outbound flow in period t respectively.

[0056] For time period t, the output constraint in the three-level constraint is converted into the outflow flow Q of the hydropower station in time period t t Constraints U3:

[0057] U3=U 3,1 ∪U 3,2

[0058]

[0059]

[0060]

[0061]

[0062] Where U 3,1 , U 3,2 Represents two subsets of U3; N t,min 、N t,max They represent the lower and upper limits of the hydropower station output during period t; f NQ (·) represents the output function relationship of the hydropower station; They represent the two outflow flow values ​​corresponding to the lower limit of the hydropower station output (the hydropower station output function relationship f NQ (·) Usually parabolic, 1 output value corresponds to 2 outflows); They respectively represent the two outflow flow values ​​corresponding to the upper limit of the hydropower station output.

[0063] ② Couple the constraints obtained in step ①. Specifically, transform the constraints of different levels for different variables into the outbound flow Q in period t. t The constraints {U1,U2,U3}, let the set W1=U1,W2=U1∩U2,W3=U1∩U2∩U3;

[0064] Initial storage capacity V t Can be changed within a given range, traversing V t The possible values ​​of {W1, W2, W3} are calculated. If all W3 are empty sets, it is considered that the three-level constraint cannot be satisfied. Otherwise, the outbound flow Q in period t is t The feasible domain W=f(L0,L t )=W3; if all W2 and W3 are empty sets, it is considered that the secondary constraint cannot be satisfied, and the feasible domain W=f(L0,L t )=W1, otherwise the feasible domain W=f(L0,L t )=W2.

[0065] ③ Repeat steps ① and ② to calculate and store the feasible domain of outbound flow under different time periods, different inbound flow rates and initial storage capacity of the time period;

[0066] Let time period t = 1, given the scheduling constraint L t , inbound flow I t Discretize within the possible range, the number of discrete points is N, and the initial storage capacity of the time period is V t Discretize within the possible range, the number of discrete points is M; according to steps ① and ②, calculate the different I in time period t t and V t Combined outbound flow Q t The feasible domain W is stored in the database; let t = t + 1, and repeat the above steps until t is greater than the number of scheduling period time periods T.

[0067] (3) Algorithms (dynamic programming algorithms, intelligent algorithms, etc.) optimize the average outbound flow Q during period t t When the time period t and the inflow flow I t and initial storage capacity V t , query the feasible domain of the corresponding decision variables in the database, and search only within the feasible domain, which significantly improves the computational efficiency of the algorithm.

[0068] In this embodiment, a maximum power generation model of a hydropower station group and a minimum flood discharge model of a hydropower station group are constructed respectively to carry out a dimensionality reduction method analysis based on a feasible domain search mapping, and the objective functions are:

[0069]

[0070] Where, I is the number of hydropower stations involved in the calculation; i is the station number; T is the total number of time periods; t is the time period number; N i,t K is the output of hydropower station i during period t; i is the output coefficient of the i hydropower station; q i,t is the power generation flow of hydropower station i in period t; H i,t is the generating head of the hydropower station i during period t;

[0071]

[0072] In the formula, S i,t is the flood discharge of hydropower station i in period t.

[0073] Taking the cascade power stations consisting of six hydropower stations A, B, C, D, E, and F in the upper reaches of the Yangtze River as the object, two different examples are designed to verify the effectiveness of the dimensionality reduction method for optimizing the scheduling of hydropower stations based on feasible domain search mapping provided in this embodiment. The six hydropower stations are series reservoirs, and from upstream to downstream they are A power station, B power station, C power station, D power station, E power station, and F power station. Except for D and F power stations, the regulation capacity of the remaining power stations is relatively strong.

[0074] In the first example, the water level in a typical year is used as input, the year is used as the scheduling period, the ten-day period is used as the calculation period, the maximum cascade power generation is used as the goal, and the stepwise optimization algorithm (POA) is selected as the solution algorithm. The calculation results before and after the dimension reduction method of this embodiment is applied are compared with the results based on CPU multi-core parallelism.

[0075] The second example uses the water inflow from July 21 to August 31 in a typical flood season as input, takes the day as the calculation period, takes the minimum cascade discharge as the goal, selects POA as the solution algorithm, and compares the calculation results before and after the dimensionality reduction method is applied, as well as the results based on CPU multi-core parallelism. The calculation program is written in JAVA and the calculation is completed on an Intel Core i7-12700 workstation with 12 cores and 64GB of memory.

[0076] In Example 1, the water level constraints of each hydropower station are set according to the reservoir dispatching regulations, the output constraints are set according to the actual dispatching situation in the past five years, and the outflow constraints are set according to the downstream ecological, water supply, irrigation and other requirements. The minimum outflow requirement for power station A is 1160m 3 / s; the minimum outflow rate of power station B is 1260m 3 / s; the minimum outflow flow of power station C is 1200m 3 / s; the minimum outflow rate of power station D is 1700m 3 / s; the minimum outflow rate of E power station from mid-September to October is 8000m 3 / s, other periods are 6000m 3 / s; the minimum outflow rate of F power station is 5500m 3 / s. Taking the water supply in a typical year as input, the water supply in 10 days does not meet the minimum outflow flow requirements of power stations A, B, C, and F, the water supply in 70 days does not meet the minimum outflow flow requirements of power station D, and the water supply in 50 days does not meet the minimum outflow flow requirements of power station E. The problem of optimizing the distribution of water between cascade power stations is more prominent. The optimization model is solved by POA. Table 1 is a comparison of the calculation results before and after the application of the dimensionality reduction method of this embodiment and the CPU multi-core calculation results. It can be seen from Table 1 that the introduction of the dimensionality reduction method and CPU parallelism did not change the annual power generation of the optimized cascade power station. When the discrete precision is low, the acceleration effect of the dimensionality reduction method of this embodiment is not as good as that of CPU parallelism. When the discrete step length is 1m, the acceleration ratio is only 4.7; when the discrete precision is high, the acceleration effect is significantly better than that of CPU parallelism. When the discrete step length is 0.1m, the acceleration ratio reaches 14.7.

[0077] Table 1: Comparison of the results of Example 1 before and after the application of dimensionality reduction methods and parallel computing

[0078]

[0079]

[0080] In Example 2, the water level constraints of each hydropower station are set according to the reservoir dispatching regulations and dispatching orders, the output constraints are set according to the actual dispatching conditions in the same period of the past five years, and the flow constraints are set according to the downstream flood control, ecology, water supply and other requirements. Taking the water inflow from July 21 to August 31 in a typical flood season as input, the water inflow exceeds the rated flow of power station A for 35 days, the water inflow exceeds the rated flow of power station B for 24 days, the water inflow exceeds the rated flow of power station C for 42 days, the water inflow exceeds the rated flow of power station D for 41 days, the water inflow exceeds the rated flow of power station E for 29 days, and the water inflow exceeds the rated flow of power station F for 42 days. The problem of optimizing the allocation of flood control storage capacity between cascade power stations is more prominent. Using the POA solution model, Table 2 is a comparison of the calculation results before and after the application of the dimensionality reduction method of this embodiment and the CPU multi-core calculation results. It can be seen from Table 2 that the introduction of the dimensionality reduction method and CPU parallelism did not change the flood discharge of the optimized cascade power station. When the discrete precision is low, the acceleration effect of the dimensionality reduction method of this embodiment is not as good as CPU parallelism. When the discrete step length is 1m, the acceleration ratio is only 4.1; when the discrete precision is high, the acceleration effect is significantly better than CPU parallelism. When the discrete step length is 0.1m, the acceleration ratio reaches 17.2.

[0081] Table 2: Comparison of the results of Example 2 before and after the application of dimensionality reduction methods and parallel computing

[0082]

[0083] Through the above example analysis, it can be concluded that the improvement of the computational efficiency of the method in this embodiment is not obvious when the discrete precision is low, and the acceleration effect is significantly better than CPU multi-core parallel when the discrete precision is high. The main reasons are as follows: when the discrete precision is low, the number of scheduling processes obtained by discrete is small, and the number of scheduling processes that do not meet the constraints that can be excluded by the method in this embodiment is limited, so the efficiency improvement is not obvious; when the discrete precision is high, the number of scheduling processes obtained by discrete is large, and a large number of scheduling processes that do not meet the constraints can be excluded by the method in this embodiment, ensuring that the search is performed within the feasible domain, so the computational efficiency is significantly optimized. Since the dimensionality reduction method is usually applied to the optimization scheduling problem of hydropower station groups with high discrete precision and complex constraints, the application advantage of the method in this embodiment is obvious.

[0084] The above embodiments are used to illustrate the present invention rather than to limit the present invention. Any equivalent changes made to the present invention within the protection scope of the claims of the present invention shall fall within the protection scope of the present invention.

Claims

1. A dimensionality reduction method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping, characterized in that: Here are the steps: (1) The dispatching and operation constraints of hydropower stations in different periods are sorted and classified, and the constraints involved in the operation of hydropower stations are divided into three levels; (2) Determine the decision variables for hydropower station dispatching under different periods and conditions, that is, the mapping relationship between the feasible domain of outflow flow and known operating conditions and dispatching constraints; (3) When optimizing the outflow of a hydropower station in a certain period of time, the algorithm can query the feasible domain of the corresponding decision variables in the database according to the period of time, the inflow and the initial storage capacity of the period, and search for the optimal solution only within the feasible domain.

2. The method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping according to claim 1 is characterized in that: In step (1), the first-level constraints are the constraints that need to be met to ensure the safe and stable operation of the reservoir, including water level constraints and storage capacity constraints; the second-level constraints are the constraints on outflow from flood control, ecology, water supply, and shipping; and the third-level constraints are the requirements of the power generation department for the output of the hydropower station.

3. The method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping according to claim 2 is characterized in that: In step (2), for time period t, the inflow flow I of the hydropower station in this period is known t and initial storage capacity V t , the outflow flow Q of the hydropower station in the previous period t-1 and output N t-1 , recorded as the known operating condition L0; The set of constraints in time period t is collectively recorded as scheduling constraint L t , the feasible domain W of the decision variables for the hydropower station dispatching in period t and L0, L t The mapping relationship between them is recorded as W = f(L0,L t ), the calculation process of the mapping relationship is as follows: ① Given the inflow and initial storage capacity of a hydropower station in a certain period, the constraints of different levels in the period are converted into constraints for the outflow in the period; ② Couple the constraints obtained in step ① to obtain the feasible domain of the outbound flow during this period; ③ Repeat steps ① and ② to calculate and store the feasible domain of outbound flow under different time periods, different inbound flow rates and initial storage capacity of the time period.

4. The method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping according to claim 3 is characterized in that: In step ①, specifically, For time period t, based on water balance, the first-level constraint is converted into the outflow flow Q of the hydropower station in time period t t Constraints U1: In the formula, Q t represents the average outflow of the hydropower station during period t; They represent the lower and upper limits of the outbound flow corresponding to the first-level constraint; V t+1,min 、V t+1,max They represent the lower and upper limits of storage capacity at the end of period t respectively; Δt represents the time period length.

5. The method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping according to claim 4 is characterized in that: In step ①, for time period t, the secondary constraint is taken as the outflow flow Q of the hydropower station during time period t t Constraints U2: In the formula, They represent the lower and upper limits of the outbound flow corresponding to the secondary constraints respectively; Q t-1 represents the average outflow of the hydropower station during period t-1; ΔQ t,re , ΔQ t,add They represent the maximum increase and decrease of outbound flow in period t respectively.

6. The method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping according to claim 5 is characterized in that: In step ①, for time period t, the three-level constraints are converted into the outflow flow Q of the hydropower station during time period t: t Constraints U3: U3=U 3,1 ∪U 3,2 Where U 3,1 , U 3,2 Represents two subsets of U3; N t,min 、N t,max They represent the lower and upper limits of the hydropower station output during period t respectively; f NQ (·) represents the output function relationship of the hydropower station; They represent the two outflow flow values ​​corresponding to the lower limit of the hydropower station output; They respectively represent the two outflow flow values ​​corresponding to the upper limit of the hydropower station output.

7. The method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping according to claim 6 is characterized in that: In step ②, specifically, the constraints of different levels for different variables are converted into the outbound flow Q in period t t The constraints {U1,U2,U3}, let the set W1=U1,W2=U1∩U2,W3=U1∩U2∩U3; Initial storage capacity V t Can be changed within a given range, traversing V t The possible values ​​of {W1, W2, W3} are calculated. If all W3 are empty sets, it is considered that the three-level constraint cannot be satisfied. Otherwise, the outbound flow Q in period t can be obtained. t The feasible domain W=f(L0,L t )=W3; if all W2 and W3 are empty sets, it is considered that the secondary constraint cannot be satisfied, and the feasible domain W=f(L0,L t )=W1, otherwise the feasible domain W=f(L0,L t )=W2.

8. The method for optimizing the scheduling of a hydropower station group based on feasible domain search mapping according to claim 7 is characterized in that: In step ③, specifically: Let time period t = 1, given the scheduling constraint L t , inbound flow I t Discretize within the possible range, the number of discrete points is N, and the initial storage capacity of the time period is V t Discretize within the possible range, the number of discrete points is M; calculate different I in time period t t and V t Combination Q t The feasible domain W is stored in the database; let t = t + 1, and repeat the above calculation steps until t is greater than the number of scheduling period time periods T.

Images