Reservoir optimization scheduling method oriented to scheduling plan

Abstract

The invention provides a reservoir optimization scheduling method for a scheduling plan, and relates to the technical field of reservoir multi-task coordination, and the method comprises the steps: randomly generating a plurality of groups of monthly power generation plans; calculating the initial reservoir capacity of the reservoir to obtain the available power generation water quantity of the reservoir; a Newton interpolation method and an improved Powell method are used for dynamically calculating the power generation water volume; the ecological water supply amount, the urban water supply amount and the agricultural water supply amount are calculated; calculating an objective function of the power generation plan meeting the constraint condition; in response to a water shortage target equal time comparison power generation target, obtaining each generation of global optimal and individual optimal power generation plans by comparing target function values; and for reaching the maximum number of iterations, obtaining an optimal power generation plan meeting constraint conditions based on each generation of global optimal and individual optimal power generation plans. According to the invention, the problem of mismatching between the power generation plan and the water supply plan in actual scheduling management is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of reservoir multi-task coordination, and in particular relates to a reservoir optimization scheduling method oriented to scheduling plan. Background Art

[0002] With the acceleration of economic and social development and urbanization, the contradiction between regional water supply and demand has become increasingly prominent. How to coordinate multiple tasks of reservoirs has become a key issue in dispatching and operation. Dispatching plans generally include power generation plans and water supply plans. In actual management, power generation plans and water supply plans are usually responsible for different departments. The former is usually issued by the power grid, and the latter is formulated by the water administration department. It is difficult to fully consider the competitive relationship between power generation and water supply. For example, an unreasonable power generation plan may lead to the failure to meet the water supply plan, resulting in certain risks in the implementation of the dispatching plan. Therefore, it is very important to comprehensively consider the competitive relationship between power generation and water supply in the total water use and process, and formulate a reasonable reservoir dispatching plan.

[0003] To optimize the matching between the reservoir power generation plan and the water supply plan, it is necessary to fully combine the differences and commonalities of power generation and water supply. In terms of differences, power generation scheduling has autonomous flexibility, with the goal of obtaining maximum benefits under the same water inflow conditions, so it tends to operate at high water levels, thereby achieving greater power generation benefits with less water discharge. Therefore, when the reservoir water storage is insufficient, the power generation water volume will be reduced. Reservoir water supply scheduling has a deterministic goal, and its water supply process is determined by the user's water demand process. Water is supplied according to water storage requirements to ensure water safety. In terms of commonalities, both need to control the reservoir water storage to be too high or too low, causing damage or loss of benefits. On the one hand, it is necessary to consider the subsequent period effect to avoid damage caused by too little water storage, and on the other hand, it is necessary to avoid the loss of benefits caused by abandonment of water due to excessive water storage. Therefore, the core of scheduling plan formulation is to reasonably allocate the water supply process in combination with the characteristics of water inflow, so that the reservoir water storage can reasonably decline during the scheduling period, and minimize the abandonment of water and water supply damage. Therefore, the difficulty in formulating a reservoir scheduling plan lies in how to comprehensively consider the differences and commonalities between power generation and water supply processes, and coordinate the relationship between the two in terms of total water use, water use process and water storage process.

[0004] According to the principle of "power dispatch is subject to water dispatch", the matching of power generation plan and water supply plan is analyzed. When the social and economic water use is damaged according to the implementation of the power generation plan, the power generation plan needs to be adjusted. The key to coordinating power generation and water supply is to adjust the power generation process. When the water supply plan cannot be met, the power generation plan of the current period can be adjusted to directly increase the amount of water discharged during the water shortage period, or the power generation plan of the previous period can be adjusted. On the one hand, the water consumption of the previous power generation can be reduced and the water supply in the subsequent period can be increased. At the same time, a certain head difference can be maintained to reduce the loss of power generation benefits. Therefore, how to determine the adjustment period and adjustment amount of the power generation plan is the difficulty of this method.

[0005] In the prior art, the PA-DDS algorithm is introduced into the multi-objective optimization model considering water supply and power generation, and the reservoir dispatching diagram is optimized with the maximum water supply benefit and the maximum power generation as the objective function. The research results are applicable to the reservoir engineering planning and dispatching rule formulation stage, but there is a lack of research on how to coordinate the power generation plan and the water supply plan in the actual dispatching management. Therefore, the optimal dispatching of reservoirs needs to be based on actual needs, with the goal of reducing the impact of the dispatching plan, analyzing and optimizing the dispatching method, improving the matching of power generation and water supply in the total water use and process, and supporting the formulation of reservoir dispatching plans and actual operations and other related work. The shortcomings are as follows:

[0006] Reservoir optimization scheduling mainly focuses on how to maximize the power generation benefits of the reservoir. There is a lack of research on how to coordinate power generation plans and water supply plans in actual scheduling management, and the synergy of scheduling plans in actual reservoir scheduling management is not considered. Summary of the invention

[0007] In view of the above-mentioned deficiencies in the prior art, the present invention provides a reservoir optimization scheduling method oriented to scheduling plan, so as to improve the matching between the power generation plan and the water supply plan in terms of the total water consumption and the water consumption process on the basis of slightly adjusting the original plan as much as possible.

[0008] In order to achieve the above purpose, the technical solution adopted by the present invention is: a reservoir optimization scheduling method oriented to scheduling plan, comprising the following steps:

[0009] S1. Randomly generate multiple groups of monthly power generation plans;

[0010] S2. Obtain the amount of water available for power generation in the reservoir by calculating the initial storage capacity of the reservoir;

[0011] S3. Based on the available water volume for power generation in the reservoir and the power generation plan, the Newton interpolation method and the improved Powell algorithm are used to dynamically calculate the power generation water volume;

[0012] S4. Based on the amount of water for power generation, calculate the ecological water supply, urban water supply and agricultural water supply respectively;

[0013] S5. Calculate the objective function of the power generation plan that meets the constraint conditions according to the calculated ecological water supply, urban water supply and agricultural water supply;

[0014] S6, determine whether the maximum calculation period has been reached, if so, proceed to step S7, otherwise, return to step S2;

[0015] S7, based on the principle that power dispatching is subject to water dispatching, when the water shortage targets are equal, the power generation targets are compared, and the global optimal and individual optimal power generation plans of each generation are obtained by comparing the objective function values ​​of step S5;

[0016] S8. Determine whether the maximum number of iterations has been reached. If so, obtain the optimal power generation plan that meets the constraints based on the global optimal and individual optimal power generation plans of each generation. Otherwise, update the power generation plan based on the global optimal and individual optimal power generation plans of each generation, with the evolution speed and direction, as well as the evolution power generation plan, and return to step S2 until the maximum number of iterations is reached.

[0017] The beneficial effect of the present invention is that by substituting the power generation plan into the reservoir water supply process to screen the optimal power generation plan that meets the constraints, the matching between the power generation plan and the water supply plan in terms of total water consumption and water use process is improved on the basis of adjusting the original plan as little as possible.

[0018] Furthermore, the expression of the amount of water available for power generation in the reservoir is as follows:

[0019] WSele t =V t -L t -Ve min

[0020]

[0021] V t =V t-1 +Win t

[0022] Among them, WSele t V represents the amount of water available for power generation in the reservoir during period t, t represents the initial storage capacity at time period t, L t Represents the amount of water loss from the reservoir during period t, Ve min Represents the limited storage capacity for power generation water withdrawal, V t-1 represents the final storage capacity of the t-1 period, α represents the reservoir leakage coefficient, V min Indicates the dead storage capacity of the reservoir, Win t Represents the amount of water entering the reservoir during period t.

[0023] The beneficial effect of the above further scheme is that the present invention avoids the reservoir water storage capacity exceeding the power generation limit storage capacity after power generation by calculating the water consumption available for power generation.

[0024] Furthermore, step S3 includes the following steps:

[0025] S301, calculating Newton interpolation polynomial difference quotient;

[0026] S302, deriving a functional relationship between water level and reservoir capacity based on Newton interpolation polynomial difference quotient;

[0027] S303. Based on the functional relationship between the water level and the reservoir capacity, the improved Powell algorithm is used to dynamically calculate the water volume for power generation.

[0028] The beneficial effects of the above further scheme are: the power generation water volume calculated by the present invention depends on multiple factors such as the power generation plan in each time period, the reservoir water storage capacity, the water storage level, the storage capacity curve, etc., wherein the Newton interpolation method is used to solve the corresponding water level of the known storage capacity; the power generation tail water level is a function of the power generation water volume, and in order to solve the higher-order equation, the improved Powell method is used to calculate the power generation water volume.

[0029] Furthermore, the expression of the Newton interpolation polynomial difference quotient is as follows:

[0030]

[0031] Among them, f[V 0 ,V 1 ,L,V n ] indicates that the function f(V) is at point V 0 ,V 1 ,L,V n The n-order difference quotient at V n represents the storage capacity data at location n, f[V 0 ,V 1 ,L,V n-1 ] indicates that the function f(V) is at point V 0 ,V 1 ,L,V n-1 The n-1 order difference quotient at n-1 Represents the storage capacity data at n-1, V 0 Indicates the first storage capacity data, f[V 1 ,L,V n-1 ,V n ] indicates that the function f(V) is at point V 1 ,L,V n-1 ,V n The n-order difference quotient at

[0032] The functional relationship between the water level and the reservoir capacity is as follows:

[0033] f(V)=f(V 0 )+(VV 0 )f[V 0 ,V 1 ]+(VV 0 )(VV 1 )f[V 0 ,V 1 ,V 2 ]

[0034] +,L,+(VV 0 )(VV 1 ),L,(VV n-1 )f[V 0,V 1 ,L,V n ]

[0035] c 0 =f(V 0 )

[0036]

[0037] Among them, f(V) represents the function of water level with respect to reservoir capacity, f(V 0 ) represents the water level corresponding to the first storage capacity data, V represents the reservoir capacity, V n-1 represents the storage capacity data at n-1, c 0 、c 1 、c 2 represents the coefficients of Newton interpolation polynomial, f(V 1 ) represents the water level corresponding to the reservoir capacity data at sequence 1, V 1 represents the storage capacity data at sequence 1, f[V 0 ,V 1 ] indicates that the function f(V) is at point V 0 ,V 1 The first-order difference quotient at f(V 2 ) represents the water level corresponding to the reservoir capacity data at sequence 2, V 2 represents the storage capacity data at sequence 2, f[V 0 ,V 2 ] indicates that the function f(V) is at point V 0 ,V 2 The first-order difference quotient at 0 ,V 1 ,V 2 ] indicates that the function f(V) is at point V 0 ,V 1 ,V 2 The second-order difference quotient at .

[0038] The beneficial effect of the above further scheme is: the use of Newton interpolation method to solve the functional relationship between water level and reservoir capacity can efficiently utilize existing data points, gradually construct polynomials through differential calculation, achieve high-precision fitting, and facilitate the dynamic update of new data points to ensure the accuracy of the water level-reservoir capacity relationship.

[0039] Furthermore, based on the functional relationship between the water level and the reservoir capacity, the improved Powell algorithm is used to dynamically calculate the power generation water volume, which is specifically:

[0040] Initialization operation: Select the initial estimate Q Et0 Q Et Iterate; set the solution accuracy ε; set the initial direction vector d 0 , where Q Etrepresents the amount of water generated during period t, Q Et0 represents the initial estimate of the amount of water required for power generation;

[0041] Constructing the objective function: Based on the amount of water generated, the objective function is constructed using the following formula, where the constructed objective function is to find the value that makes F(Q Et(k+1) ) is closer to 0 Q Et(k+1) :

[0042] F(Q Et(k+1) )=Q Et(k) -Q Et(k+1)

[0043] Among them, Q Et(k) represents the amount of water generated in the kth iteration of period t, F(Q Et(k+1) ) represents the objective function value of the dynamic solution of the power generation water quantity in the k+1 iteration of the time period t, Q Et(k+1) represents the amount of water generated in the k+1 iteration of period t;

[0044] Linear search: along direction d k Perform a one-dimensional search to find a new point Q Et(k+1) So that F(Q Et(k+1) ) is close to 0;

[0045] Update direction: Based on the linear search results, determine the direction d k value;

[0046] Check convergence: Determine the new function value F(Q Et(k+1) ) is less than the precision ε. If so, stop the iteration and output Q based on the updated result. Et(k+1) As the solution, that is, the amount of water generated, otherwise, return to the linear search step.

[0047] The beneficial effect of the above further scheme is that the present invention realizes iterative evolution of the power generation water volume, determines the power generation water volume that meets the power generation plan, and solves the objective function F(Q Et(k+1) ) value is used to determine whether the judgment accuracy of ending the iteration has been reached, and the required amount of water for power generation can be solved based on the power generation plan.

[0048] Furthermore, the linear search expression is as follows:

[0049] Q Et(k+1) =Q Et(k) +ad k

[0050]

[0051] Among them, Q Et(k+1) represents the amount of water generated in the k+1th iteration of period t, Q Et(k)represents the amount of water generated in the kth iteration of period t, a represents the search step length, d k represents the search direction of the kth iteration, E t represents the planned power generation during period t, V t represents the initial storage capacity at time period t, f(V t ) represents the functional relationship between water level and reservoir capacity, g(Q Et ) represents the functional relationship between the tail water level and the amount of water generated, and Δh represents the head loss.

[0052] Furthermore, the expression of the update direction is as follows:

[0053]

[0054] Among them, Q Et(k+1) represents the amount of water generated in the k+1th iteration of period t, Q Et(k) represents the amount of water generated in the kth iteration of period t, d k Indicates the search direction for the kth iteration.

[0055] Furthermore, the expression of the ecological water supply is as follows:

[0056]

[0057] Among them, WSe t represents the ecological water supply in period t, Q Et represents the amount of water generated during period t, V t represents the initial storage capacity at time period t, Ve min Indicates the limited storage capacity for power generation water withdrawal, Dec t represents the planned ecological water supply in period t, Weav t It represents the actual water supply after the reservoir generates electricity during period t;

[0058] The expression of town water supply is as follows:

[0059]

[0060] Among them, WSt t represents the urban water supply during period t, Qtav t Dt represents the remaining amount of water available to towns from tail water of power generation in period t after the ecological water supply is updated. t represents the planned urban water supply volume during period t, Wen t The amount of water that can still be stored in the town after the water storage reaches the water withdrawal limit for power generation in period t;

[0061] The expression of agricultural water supply is as follows:

[0062]

[0063] Among them, WSa t represents the agricultural water supply during period t, Qarav t It represents the remaining amount of water for agricultural irrigation after the power generation tail water in period t after the update, Dar t represents the planned agricultural water supply during period t, Wtow t It indicates the amount of water that can still be used for agriculture after the water storage capacity reaches the water withdrawal limit for power generation during period t.

[0064] The beneficial effect of the above further scheme is that this process realizes the allocation of power generation tail water according to the priorities of ecology, town and agriculture, avoiding the allocation of water exceeding the tail water volume or the planned volume.

[0065] Furthermore, the objective function in step S5 includes:

[0066] The objective function is to minimize ecological water shortage, urban water shortage and agricultural water shortage:

[0067]

[0068] Where WDe represents the water shortage of water users, i represents the classification code of water users, where ecological users in the river are 1, urban users are 2, and agricultural users are 3. it represents the planned water supply volume of water user i in period t, Wus it represents the water supply of water user i in period t;

[0069] The objective function is to minimize the deviation of power generation plan:

[0070]

[0071] Among them, R represents the deviation of power generation plan, which is expressed by the sum of squares of the difference between actual power generation and planned power generation as a proportion of planned power generation, Na t represents the actual power generation during period t, E t represents the planned power generation during period t.

[0072] The beneficial effect of the above further scheme is that the present invention determines the objective function from two directions, namely the water supply plan and the power generation plan, combines with the particle swarm algorithm to determine the global optimal solution and the individual optimal solution, and iterates the power generation plan.

[0073] Furthermore, the expressions of the evolution speed and direction in step S7 are as follows:

[0074] v i'd (t'+1)=wv i'd (t')+c 1 rand()(p id -x i'd(t'))+c 2 rand()(p gd -x i'd (t'))

[0075] The expression of the evolutionary power generation plan is as follows:

[0076] x i'd (t'+1)=x i'd (t')+v i'd (t'+1)

[0077] Among them, v i'd (t'+1) represents the velocity component of the ith particle in the dth dimension in the t'+1th iteration, w represents the inertia weight, v i'd (t') represents the velocity component of the ith particle in the dth dimension in the t'th iteration, c 1 and c 2 Both represent learning factors, rand() represents a random number between 0 and 1, and p id represents the individual optimal particle, p gd is the global optimal particle, x i'd (t') represents the position component of the ith particle in the dth dimension in the t'th iteration, x i'd (t'+1) represents the position component of the ith particle in the dth dimension in the t'+1th iteration, where the particle is a power generation plan.

[0078] The beneficial effect of the above further scheme is that the present invention optimizes the scheduling of reservoirs based on the scheduling plan, takes the water shortage of water users and the minimum deviation from the power generation plan as the goal based on the algorithm (such as the particle swarm algorithm), takes the monthly power generation plan as the decision variable, and determines the final power generation plan through iterative calculation, thereby improving the matching of power generation and water supply in the total water consumption and process. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0080] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0081] Example

[0082] There are two main modes of reservoir power generation and water supply. One is the sharing of power generation and water supply processes, and there is no overlapping relationship between the two water volumes. The other is the coupling of power generation and water supply processes, and the tail water of power generation is used for urban water supply, agricultural irrigation, river ecology and other users. Both combination methods need to consider the matching of power generation plan and water supply plan in terms of total amount and process, but relatively speaking, the coupling mode of power generation and water supply has the dual relationship of water use competition and water use coordination, and the factors affecting the transfer of power generation and water supply benefits are more complex. Therefore, the present invention is mainly based on the coupling mode of power generation and water supply processes, and analyzes the reservoir optimization scheduling method for scheduling plans. The present invention is used in the mode of coupling power generation and water supply processes, and the tail water of power generation is used for urban water supply, agricultural irrigation, river ecology and other users. Based on actual needs, the present invention aims to reduce the impact of scheduling plans, analyze and optimize scheduling methods, improve the matching of power generation and water supply in total water use and process, and support related work such as reservoir scheduling plan formulation and actual operation. Figure 1 As shown, the present invention provides a reservoir optimization scheduling method oriented to scheduling plan, and its implementation method is as follows:

[0083] S1. Randomly generate multiple groups of monthly power generation plans;

[0084] In this embodiment, n groups of monthly power generation plans X are randomly generated. n ={x n1 ,x n2 ,…,x n12}(n=1,2,3,...,N), where x n12 represents the power generation plan of the nth particle in the 12th month, and N represents the total power generation plan group.

[0085] S2. Obtain the amount of water available for power generation in the reservoir by calculating the initial storage capacity of the reservoir;

[0086] In this embodiment, the initial storage capacity of the reservoir is calculated as shown in formula (1):

[0087] V t =V t-1 +Win t (1)

[0088] Calculate the water loss of the reservoir, see formula (2):

[0089]

[0090] Calculate the amount of water available for power generation in the reservoir, see formula (3):

[0091] WSele t =V t -L t -Ve min (3)

[0092] Among them, WSele t V represents the amount of water available for power generation in the reservoir during period t, t represents the initial storage capacity at time period t, L t Represents the amount of water loss from the reservoir during period t, Ve min Represents the limited storage capacity for power generation water withdrawal, V t-1 represents the final storage capacity of the t-1 period, α represents the reservoir leakage coefficient, V min Indicates the dead storage capacity of the reservoir, Win t Represents the amount of water entering the reservoir during period t.

[0093] S3. Based on the available water volume for power generation in the reservoir and the power generation plan, the Newton interpolation method and the improved Powell algorithm are used to dynamically calculate the power generation water volume. The implementation method is as follows:

[0094] S301, calculating Newton interpolation polynomial difference quotient;

[0095] S302, deriving a functional relationship between water level and reservoir capacity based on Newton interpolation polynomial difference quotient;

[0096] S303, based on the functional relationship between water level and reservoir capacity, the improved Powell algorithm is used to dynamically calculate the amount of water for power generation, which is specifically:

[0097] Initialization operation: Select the initial estimate Q Et0 Q Et Iterate; set the solution accuracy ε; select the initial direction vector d 0 = -1, where Q Et represents the amount of water generated during period t, Q Et0 represents the initial estimate of the amount of water required for power generation;

[0098] Constructing the objective function: Based on the amount of water generated, construct the objective function, where the constructed objective function is to find the value that makes F(Q Et(k+1) ) is closer to 0 Q Et(k+1) , where F(Q Et(k+1) ) represents the objective function value of the dynamic solution of the power generation water quantity in the k+1 iteration of the time period t, Q Et(k+1) represents the amount of water generated in the k+1 iteration of period t;

[0099] Linear search: along direction d k Perform a one-dimensional search to find a new point Q Et(k+1) So that F(Q Et(k+1) ) is close to 0;

[0100] Update direction: Based on the linear search results, determine the direction d k value;

[0101] Check convergence: Determine the new function value F(Q Et(k+1) ) is less than the precision ε. If so, stop the iteration and output Q based on the updated result. Et(k+1) As the solution, that is, the amount of water generated, otherwise, return to the linear search step.

[0102] In this embodiment, the power generation water volume is calculated. The power generation water volume of the reservoir is dynamically calculated based on the reservoir water storage capacity and the power generation plan, and the power generation output formula (4) is converted into the power generation water volume calculation formula (5).

[0103] N t =AQ Et H t (4)

[0104]

[0105] Among them, N t Indicates power generation output or power generation plan; Q Et H represents the amount of water generated during period t; t represents the power generation head difference, which is calculated according to formula (6); A represents the power generation output coefficient of the power station, which is generally taken as 8.5 and can be determined based on the actual situation of the power station:

[0106]

[0107] Formula (6) is the calculation formula for generating water head: represents the average water level upstream of the hydropower station, represents the average water level downstream of the hydropower station, and Δh represents the head loss. Formula (7) is the water level storage capacity curve formula, V t is the initial reservoir capacity in period t. Formula (8) is the downstream water level and flow curve formula, Q t Represents the reservoir water volume during period t.

[0108] Substituting formula (6) to formula (8) into formula (5), we get formula (9) for calculating water consumption for power generation:

[0109]

[0110] Among them, Q Et represents the amount of water used for power generation during period t; E t represents the planned power generation during period t; V t represents the initial storage capacity during period t; Δh represents the head loss.

[0111] It can be seen from formula (9) that the amount of water for power generation depends on many factors such as the power generation plan in each period, the water storage capacity of the reservoir, the water storage level, the reservoir capacity curve, etc. The Newton interpolation method is used to solve the upstream average water level under the known reservoir capacity; the tail water level of power generation is a function of the amount of water for power generation. Therefore, there are unknown quantities of water for power generation on both sides of the equal sign in formula (9). In order to solve the higher-order equation, the improved Powell algorithm Powell is used to calculate the amount of water for power generation.

[0112] (1) Newton interpolation method solution process

[0113] The reservoir capacity data of the known water level reservoir capacity curve is: V = {V 0 ,V 1 ,V 2 ,…,V k-1 ,V k ,…,V n}, the corresponding water level data is: h = {h 0 ,h 1 ,h 2 ,…,h k-1 ,h k ,…,h n}, the functional relationship between water level and reservoir capacity is h=f(V).

[0114] 1) First calculate the Newton interpolation polynomial difference quotient and define the function f(V) at point V 0 ,V k The first-order difference quotient at point V is shown in formula (10), and the function f(V) is defined as 0 ,V 1 ,V k The second-order difference quotient at point V is shown in formula (11). Similarly, f(V) is defined at point V 0 ,V 1 ,V 2 ,…,V k-1 ,V k The n-order difference quotient at is shown in formula (12):

[0115]

[0116]

[0117] 2) Derive the functional relationship between water level and reservoir capacity. The Newton interpolation polynomial is shown in formula (13):

[0118] N n (V) = c 0 +c 1 (VV 0 )+c 2 (VV 0 )(VV 1 )+L+c n(VV 0 )(VV 1 )L(VV n-1 ) (13)

[0119] In the formula, N n (V 0 )=c 0 =f(V 0 ) to get c 0 =f(V 0 ); by N n (V 1 )=f(V 0 )+c 1 (V 1 -V 0 )=f(V 1 )get By N n (V 2 )=f(V 0 )+f[V 0 ,V 1 ](V 2 -V 0 )+c 2 (V 2 -V 0 )(V 2 -V 1 )=f(V 2 )get By analogy, all coefficients of the Newton interpolation formula can be obtained by the definition of the difference quotient and mathematical induction. Substituting them into formula (13) we can obtain the functional relationship between the water level and the reservoir capacity, as shown in formula (14):

[0120]

[0121] Among them, f[V 0 ,V 1 ,L,V n ] indicates that the function f(V) is at point V 0 ,V 1 ,L,V n The n-order difference quotient at V n represents the storage capacity data at location n, f[V 0 ,V 1 ,L,V n-1 ] indicates that the function f(V) is at point V 0 ,V 1 ,L,V n-1 The n-1 order difference quotient at n-1 Represents the storage capacity data at n-1, V 0 Indicates the first storage capacity data, f[V 1 ,L,V n-1,V n ] indicates that the function f(V) is at point V 1 ,L,V n-1 ,V n The n-order difference quotient at the reservoir, f(V) represents the function of water level with respect to reservoir capacity, f(V 0 ) represents the water level corresponding to the first storage capacity data, V represents the reservoir capacity, V 0 Indicates the first storage capacity data, f[V 0 ,V 1 ,L,V n ] indicates that the function f(V) is at point V 0 ,V 1 ,L,V n The n-order difference quotient at V n-1 represents the storage capacity data at n-1, c 0 、c 1 、c 2 represents the coefficients of Newton interpolation polynomial, f(V 1 ) represents the water level corresponding to the reservoir capacity data at sequence 1, V 1 represents the storage capacity data at sequence 1, f[V 0 ,V 1 ] indicates that the function f(V) is at point V 0 ,V 1 The first-order difference quotient at f(V 2 ) represents the water level corresponding to the reservoir capacity data at sequence 2, V 2 represents the storage capacity data at sequence 2, f[V 0 ,V 2 ] indicates that the function f(V) is at point V 0 ,V 2 The first-order difference quotient at 0 ,V 1 ,V 2 ] indicates that the function f(V) is at point V 0 ,V 1 ,V 2 The second-order difference quotient at .

[0122] (2) Improved Powell method solution process

[0123] 1) Initialization: ① Choose a reasonable initial estimate Q Et0 Q Et Iterate; ② Set the solution accuracy ε; ③ Select the initial direction vector d 0 =-1, Q Et represents the amount of water generated during period t, Q Et0 Represents the initial estimate of the amount of water for power generation, which can be the available water.

[0124] 2) Constructing the objective function: The objective function is defined as shown in formula (15). The constructed objective function is to find the value that makes F(Q Et(k+1) ) is closer to 0:

[0125] F(Q Et(k+1) )=Q Et(k) -Q Et(k+1) (15)

[0126] Among them, F(Q Et ) represents the objective function for dynamically solving the power generation water quantity, E t represents the planned power generation in period t, Δh represents the head loss, A represents the power generation coefficient of the power station, f(V t ) represents the functional relationship between water level and reservoir capacity, represents the average water level upstream of the hydropower station, g(Q Et ) represents the functional relationship between the tail water level and the amount of water generated.

[0127] 5) Linear search: along direction d k Perform a one-dimensional search to find a new point Q Et(k+1) So that F(Q Et(k+1) ) is closer to 0:

[0128] Q Et(k+1) =Q Et(k) +ad k (16)

[0129]

[0130] Among them, Q Et(k+1) represents the amount of water generated in the k+1th iteration of period t, Q Et(k) represents the amount of water generated in the kth iteration of period t, a represents the search step length, d k represents the search direction of the kth iteration, E t represents the planned power generation during period t, V t represents the initial storage capacity at time period t, f(V t ) represents the functional relationship between water level and reservoir capacity, g(Q Et ) represents the functional relationship between the tail water level and the amount of water generated, and Δh represents the head loss.

[0131] 6) Update direction: Calculate the new function value F(Q Et(k+1) ), the direction update formula is shown in formula (18):

[0132]

[0133] Among them, Q Et(k+1) represents the amount of water generated in the k+1th iteration of period t, Q Et(k)represents the amount of water generated in the kth iteration of period t, d k Indicates the search direction for the kth iteration.

[0134] 7) Check convergence: If F(Q Et(k+1) )<ε, then stop the iteration and output Q Et(k+1) As a solution.

[0135] S4. Based on the amount of water for power generation, calculate the ecological water supply, urban water supply and agricultural water supply respectively;

[0136] (1) Ecological water supply

[0137] The tailwater of power generation is used to ensure the downstream ecological flow first. However, when the water storage capacity of the reservoir reaches the restricted storage capacity of power generation water withdrawal, the water storage capacity exceeding the dead storage capacity can continue to supply the ecology, as shown in formula (19):

[0138]

[0139] Among them, WSe t represents the ecological water supply in period t, Q Et represents the amount of water generated during period t, V t represents the initial storage capacity at time period t, Ve min Indicates the limited storage capacity for power generation water withdrawal, Dec t represents the planned ecological water supply in period t, Weav t It indicates the actual water supply after the reservoir generates electricity during period t.

[0140] (2) Urban water supply

[0141] Considering the importance of users outside the river, the water supply is calculated in the order of urban water supply first and agricultural irrigation second. The urban water supply calculation is shown in formula (20):

[0142]

[0143] Among them, WSt t represents the urban water supply during period t, Qtav t Dt represents the remaining amount of water available to towns from tail water of power generation in period t after the ecological water supply is updated. t represents the planned urban water supply volume during period t, Wen t It is the amount of water that can still be stored in towns after the water storage reaches the restricted storage capacity for power generation in period t.

[0144] (3) Agricultural water supply

[0145] After the urban water supply is completed, the remaining water volume of the power generation tail water is updated, and the agricultural water supply calculation is shown in formula (21):

[0146]

[0147] Among them, WSa t represents the agricultural water supply during period t, Qarav t It represents the remaining amount of water for agricultural irrigation after the power generation tail water in period t after the update, Dar t represents the planned agricultural water supply during period t, Wtow t It indicates the amount of water that can still be used for agriculture after the water storage capacity reaches the water withdrawal limit for power generation during period t.

[0148] S5. Calculate the objective function of the power generation plan that meets the constraint conditions based on the calculated ecological water supply, urban water supply and agricultural water supply;

[0149] In this embodiment, (1) the water shortage is minimal

[0150] The objective function is to minimize the water shortage in ecology, towns and agriculture.

[0151]

[0152] Where WDe represents the water shortage of water users, i represents the classification code of water users, where ecological users in the river are 1, urban users are 2, and agricultural users are 3. it represents the planned water supply volume of water user i in period t, Wus it It represents the water supply of water user i in period t.

[0153] (2) Minimum deviation from power generation plan

[0154] In order to ensure the stability of the power grid, the adjustment range of the power generation plan needs to be minimized, so the minimum deviation of the power generation plan is used as the objective function.

[0155]

[0156] Among them, R represents the deviation of power generation plan, which is expressed by the sum of squares of the difference between actual power generation and planned power generation as a proportion of planned power generation, Na t represents the actual power generation during period t, E t represents the planned power generation during period t.

[0157] S6, determine whether the maximum calculation period has been reached, if so, proceed to step S7, otherwise, return to step S2;

[0158] S7, based on the principle that power dispatching is subject to water dispatching, when the water shortage targets are equal, the power generation targets are compared, and the global optimal and individual optimal power generation plans of each generation are obtained by comparing the objective function values ​​of step S5;

[0159] In this embodiment, according to the principle of "power regulation is subject to water regulation", water shortage targets are compared first in the optimization process, and when the water shortage targets are equal, power generation targets are compared. The global optimal and individual optimal power generation plans of each generation are selected based on the comparison of objective functions.

[0160] S8. Determine whether the maximum number of iterations has been reached. If so, obtain the optimal power generation plan that meets the constraints based on the global optimal and individual optimal power generation plans of each generation. Otherwise, update the power generation plan based on the global optimal and individual optimal power generation plans of each generation, with the evolution speed and direction, as well as the evolution power generation plan, and return to step S2 until the maximum number of iterations is reached.

[0161] In this embodiment, it is determined whether the maximum number of iterations has been reached. If not, the power generation plan evolution speed and direction are updated based on the particle swarm algorithm, and the power generation plan is evolved. The speed and position update formulas are shown in formulas (24) and (25). The loop is run from step S2 until the maximum number of iterations of the particle swarm algorithm is reached. At this time, the operation is terminated and the optimal power generation plan that meets the constraints is output:

[0162] v id (t+1)=wv id (t)+c 1 rand()(p id -x id (t))+c 2 rand()(p gd -x id (t)) (24)

[0163] x id (t+1)=x id (t)+v id (t+1) (25)

[0164] Among them, v i'd (t'+1) represents the velocity component of the ith particle in the dth dimension in the t'+1th iteration, w represents the inertia weight, v i'd (t') represents the velocity component of the ith particle in the dth dimension in the t'th iteration, c 1 and c 2 Both represent learning factors, rand() represents a random number between 0 and 1, and p id represents the individual optimal particle, p gd is the global optimal particle, x i'd (t') represents the position component of the ith particle in the dth dimension in the t'th iteration, x i'd (t'+1) represents the position component of the ith particle in the dth dimension in the t'+1th iteration, where the particle is a power generation plan.

[0165] In this embodiment, other optimization methods may be used, such as a genetic algorithm, a simulated annealing algorithm, etc.

Claims

1. A reservoir optimization scheduling method oriented to scheduling plan, characterized in that: The following steps are involved: S1. Randomly generate multiple groups of monthly power generation plans; S2. Obtain the amount of water available for power generation in the reservoir by calculating the initial storage capacity of the reservoir; S3. Based on the available water volume for power generation in the reservoir and the power generation plan, the Newton interpolation method and the improved Powell algorithm are used to dynamically calculate the water volume for power generation; S4. Based on the amount of water for power generation, calculate the ecological water supply, urban water supply and agricultural water supply respectively; S5. Calculate the objective function of the power generation plan that meets the constraint conditions according to the calculated ecological water supply, urban water supply and agricultural water supply; S6, determine whether the maximum calculation period has been reached, if so, proceed to step S7, otherwise, return to step S2; S7, based on the principle that power dispatching is subject to water dispatching, when the water shortage targets are equal, the power generation targets are compared, and the global optimal and individual optimal power generation plans of each generation are obtained by comparing the objective function values ​​of step S5; S8. Determine whether the maximum number of iterations has been reached. If so, obtain the optimal power generation plan that meets the constraints based on the global optimal and individual optimal power generation plans of each generation. Otherwise, update the power generation plan based on the global optimal and individual optimal power generation plans of each generation, with the evolution speed and direction, as well as the evolution power generation plan, and return to step S2 until the maximum number of iterations is reached.

2. The reservoir optimization scheduling method oriented to scheduling plan according to claim 1 is characterized in that: The expression of the amount of water available for power generation in the reservoir is as follows: WSele t =V t -L t -And min V t =V t-1 +Win t Among them, WSele t V represents the amount of water available for power generation in the reservoir during period t, t represents the initial storage capacity at time period t, L t Ve represents the amount of water loss from the reservoir during period t, min Represents the limited storage capacity for power generation water withdrawal, V t-1 represents the final storage capacity of the t-1 period, α represents the reservoir leakage coefficient, V min Indicates the dead storage capacity of the reservoir, Win t Represents the amount of water entering the reservoir during period t.

3. The reservoir optimization scheduling method oriented to scheduling plan according to claim 1 is characterized in that: The step S3 comprises the following steps: S301, calculating Newton interpolation polynomial difference quotient; S302, deriving a functional relationship between water level and reservoir capacity based on Newton interpolation polynomial difference quotient; S303. Based on the functional relationship between the water level and the reservoir capacity, the improved Powell algorithm is used to dynamically calculate the water volume for power generation.

4. The reservoir optimization scheduling method oriented to scheduling plan according to claim 3 is characterized in that: The expression of the Newton interpolation polynomial difference quotient is as follows: Among them, f[V0,V1,L,V n ] indicates that the function f(V) is at points V0, V1, L, V n The n-order difference quotient at V n represents the storage capacity data at n locations, f[V0,V1,L,V n-1 ] indicates that the function f(V) is at points V0, V1, L, V n-1 The n-1 order difference quotient at n-1 represents the storage capacity data at n-1, V0 represents the first storage capacity data, f[V1,L,V n-1 ,V n ] indicates that the function f(V) is at the point V1,L,V n-1 ,V n The n-order difference quotient at The functional relationship between the water level and the reservoir capacity is as follows: f(V)=f(V0)+(V-V0)f[V0,V1]+(V-V0)(V-V1)f[V0,V1,V2]+,L,+(V-V0)(V-V1),L,(VV n-1 )f[V0,V1,L,V n ] c0=f(V0) Among them, f(V) represents the function of water level with respect to reservoir capacity, f(V0) represents the water level corresponding to the first reservoir capacity data, V represents the reservoir capacity, V n-1 represents the reservoir capacity data at n-1, c0, c1, c2 represent the coefficients of Newton interpolation polynomial, f(V1) represents the water level corresponding to the reservoir capacity data at sequence 1, V1 represents the reservoir capacity data at sequence 1, f[V0,V1] represents the first-order difference quotient of function f(V) at points V0 and V1, f(V2) represents the water level corresponding to the reservoir capacity data at sequence 2, V2 represents the reservoir capacity data at sequence 2, f[V0,V2] represents the first-order difference quotient of function f(V) at points V0 and V2, and f[V0,V1,V2] represents the second-order difference quotient of function f(V) at points V0,V1,V2.

5. The reservoir optimization scheduling method oriented to scheduling plan according to claim 3 is characterized in that: Based on the functional relationship between water level and reservoir capacity, the improved Powell algorithm is used to dynamically calculate the amount of water for power generation, which is specifically: Initialization operation: Select the initial estimate Q Et0 Q Et Iterate; set the solution accuracy ε; set the initial direction vector d0, where Q Et represents the amount of water generated during period t, Q Et0 represents the initial estimate of the amount of water required for power generation; Constructing the objective function: Based on the amount of water generated, the objective function is constructed using the following formula, where the constructed objective function is to find the value that makes F(Q Et(k+1) ) is closer to 0 Q Et(k+1) : F(Q Et(k+1) )=Q Et(k) -Q Et(k+1) Among them, Q Et(k) represents the amount of water generated in the kth iteration of period t, F(Q Et(k+1) ) represents the objective function value of the dynamic solution of the power generation water quantity in the k+1 iteration of the time period t, Q Et(k+1) represents the amount of water generated during the k+1th iteration of period t; Linear search: along direction d k Perform a one-dimensional search to find a new point Q Et(k+1) So that F(Q Et(k+1) ) is close to 0; Update direction: Based on the linear search results, determine the direction d k value; Check convergence: Determine the new function value F(Q Et(k+1) ) is less than the precision ε. If so, stop the iteration and output Q based on the updated result. Et(k+1) As the solution, that is, the amount of water generated, otherwise, return to the linear search step.

6. The reservoir optimization scheduling method oriented to scheduling plan according to claim 5 is characterized in that: The linear search expression is as follows: Q Et(k+1) =Q Et(k) +ad k Among them, Q Et(k+1) represents the amount of water generated in the k+1th iteration of period t, Q Et(k) represents the amount of water generated in the kth iteration of period t, a represents the search step length, d k represents the search direction of the kth iteration, E t represents the planned power generation during period t, V t represents the initial storage capacity at time period t, f(V t ) represents the functional relationship between water level and reservoir capacity, g(Q Et ) represents the functional relationship between the tail water level and the amount of water generated, and Δh represents the head loss.

7. The reservoir optimization scheduling method oriented to scheduling plan according to claim 5 is characterized in that: The expression of the update direction is as follows: Among them, Q Et(k+1) represents the amount of water generated in the k+1th iteration of period t, Q Et(k) represents the amount of water generated in the kth iteration of period t, d k Indicates the search direction for the kth iteration.

8. The reservoir optimization scheduling method oriented to scheduling plan according to claim 1 is characterized in that: The expression of the ecological water supply is as follows: Among them, WSe t represents the ecological water supply in period t, Q Et Represents the amount of water generated during period t, V t represents the initial storage capacity at time period t, Ve min Indicates the limited storage capacity for power generation water withdrawal, Dec t represents the planned ecological water supply in period t, Weav t It represents the actual water supply after the reservoir generates electricity during period t; The expression of town water supply is as follows: Among them, WSt t represents the urban water supply during period t, Qtav t Dt represents the remaining amount of water available to towns from tail water of power generation in period t after the ecological water supply is updated. t represents the planned urban water supply volume during period t, Wen t The amount of water that can still be stored in the town after the water storage reaches the water withdrawal limit for power generation in period t; The expression of agricultural water supply is as follows: Among them, WSa t represents the agricultural water supply during period t, Qarav t It represents the remaining amount of water for agricultural irrigation after the power generation tail water in period t after the update, Dar t represents the planned agricultural water supply in period t, Wtow t It indicates the amount of water that can still be used for agriculture after the water storage capacity reaches the water withdrawal limit for power generation during period t.

9. The reservoir optimization scheduling method oriented to scheduling plan according to claim 1 is characterized in that: The objective function in step S5 includes: The objective function is to minimize ecological water shortage, urban water shortage and agricultural water shortage: Where WDe represents the water shortage of water users, i represents the classification code of water users, where ecological users in the river are 1, urban users are 2, and agricultural users are 3. t represents the planned water supply volume of water user i in period t, Wus it represents the water supply of water user i in period t; The objective function is to minimize the deviation of power generation plan: Among them, R represents the deviation of power generation plan, which is expressed by the sum of squares of the difference between actual power generation and planned power generation as a proportion of planned power generation, Na t represents the actual power generation during period t, E t represents the planned power generation during period t.

10. The reservoir optimization scheduling method oriented to scheduling plan according to claim 1 is characterized in that: The expressions of evolution speed and direction in step S8 are as follows: v i'd (t'+1)=wv i'd (t')+c1rand( )(p id -x i'd (t'))+c2rand( )(p gd -x i'd (t')) The expression of the evolutionary power generation plan is as follows: x i'd (t'+1)=x i'd (t')+v i'd (t'+1) Among them, v i'd (t'+1) represents the velocity component of the ith particle in the dth dimension in the t'+1th iteration, w represents the inertia weight, v i'd (t') represents the velocity component of the ith particle in the dth dimension in the t'th iteration, c1 and c2 both represent learning factors, rand( ) represents a random number between 0 and 1, and p id represents the individual optimal particle, p gd is the global optimal particle, x i'd (t') represents the position component of the ith particle in the dth dimension in the t'th iteration, x i'd (t'+1) represents the position component of the ith particle in the dth dimension in the t'+1th iteration, where the particle is a power generation plan.

Images