Reservoir group optimization scheduling method, system, device and storage medium based on IMOSSA

Abstract

The present invention discloses a reservoir group optimization scheduling method and system, equipment and storage medium based on IMOSSA, which belongs to the field of water conservancy and hydropower. The scheduling method of the present invention includes: using the entropy weight method to screen the global optimal solution, based on the reference point strategy to perform individual screening and using the binomial crossover method to perform the finder individual update, and determining an improved multi-objective sparrow algorithm IMOSSA for solving multi-objective optimization problems; and collecting and collating the basic information of the reservoir group, generalizing the reservoir group system, and determining the hydraulic relationship between the reservoirs; and analyzing the tasks of the reservoir group, and constructing a multi-objective joint optimization scheduling mathematical model for the reservoir group; executing IMOSSA to solve the multi-objective joint optimization scheduling model of the reservoir group to obtain the Pareto non-inferior solution set, and determining the reservoir group scheduling operation plan. The reservoir group optimization scheduling method based on IMOSSA provided by the present invention can achieve global optimization, improve calculation efficiency, and meet the requirements of autonomous selection of reservoir scheduling plans.

Description

Technical Field

[0001] The invention belongs to the field of water conservancy and hydropower, and in particular to an IMOSSA-based reservoir group optimization scheduling method and system, equipment and storage medium. Background Art

[0002] A reasonable and efficient reservoir operation mode can greatly enhance the social and economic benefits. Faced with new challenges encountered in the joint operation of reservoir groups, such as the allocation of common water supply tasks in complex reservoir groups, the solution of reservoir group optimization operation and multi-objective optimization operation decision-making, how to quickly and accurately determine the optimal joint operation plan of reservoir groups through model construction and solution has become a key scientific and technical issue for the optimal allocation and real-time regulation of water resources in the basin. The multi-objective evolutionary algorithm that introduces the Pareto non-inferior solution set theory on the basis of traditional intelligent algorithms is widely used in the joint optimization operation of reservoir groups, but some intelligent optimization algorithms still have the shortcomings of large randomness in the optimization process and easy to fall into local search. Therefore, the proposed joint optimization operation modeling and efficient solution method of reservoir groups have important theoretical significance and practical value for making full use of basin water resources and improving the operation and management level of basin hydropower systems.

[0003] In recent years, with the development of intelligent optimization theory and the advancement of computer technology, scholars have begun to introduce multi-objective intelligent optimization algorithms represented by NSGA, NSGA-II, NSGA-III, etc. to solve the multi-objective scheduling problem of reservoir groups. Existing optimization scheduling algorithms mostly focus on improving the computational efficiency of the algorithm itself, and fail to conduct in-depth analysis and improvement based on instance characteristics, making it difficult to meet actual engineering needs. The Sparrow Search Algorithm (SSA) is a new intelligent optimization algorithm proposed by Xue and Shen of Donghua University in 2020. It mainly simulates the foraging and anti-predation behavior of sparrow groups, and has the characteristics of fast convergence speed, fewer parameters, and strong full set search capabilities. In the field of reservoir scheduling, literature on the application of SSA is scarce. The joint optimization scheduling of reservoir groups proposed by the multi-objective frog leaping algorithm at this stage has the defects of being prone to local optimal solutions and uneven distribution of non-inferior solutions. Summary of the invention

[0004] In view of this, the present invention proposes a reservoir group joint optimization scheduling method based on an improved multi-objective sparrow search algorithm (Improved multi-objective sparrow search algorithm, IMOSSA) to achieve efficient solution to the complex reservoir group multi-objective joint optimization scheduling problem.

[0005] The present invention is achieved through the following technical solutions:

[0006] A first aspect of the present invention provides a reservoir group joint optimization scheduling method based on IMOSSA, the method comprising:

[0007] The entropy weight method is used to screen the global optimal solution, the reference point strategy is used for individual screening, and the binomial crossover method is used to update the individual discoverer, and an IMOSSA is determined to solve the multi-objective optimization problem;

[0008] Based on the collected and sorted basic information of the reservoir group system, the reservoir group system is generalized to determine the hydraulic and hydrological relationship between the reservoirs; wherein the basic information includes runoff data, the geographical location and basic characteristic parameters of the reservoir group and the river channel, the water level-storage capacity relationship curve, and the downstream water level-discharge flow relationship curve;

[0009] Based on the hydraulic and hydrological relationship between the reservoirs, the tasks of the reservoir group are analyzed and a multi-objective joint optimization scheduling model of the reservoir group is established; wherein the objective function of the multi-objective joint optimization scheduling model of the reservoir group includes at least two of the maximum power generation, the minimum water shortage, the minimum water abandonment, the maximum minimum output, the lowest operating water level in front of the reservoir, and the minimum reservoir discharge flow, and the constraint conditions include water balance, reservoir discharge flow, unit flow capacity, downstream flood control flow, reservoir water level, and non-negative constraints;

[0010] Execute the IMOSSA to solve the multi-objective joint optimization scheduling model of the reservoir group, take the water level value / flow of each reservoir in each time period as the decision variable, determine the upper and lower limits of the water level / flow value of each reservoir, divide the cycle period of the overall scheduling of the reservoir group, obtain the Pareto non-inferior solution set, and determine the scheduling operation plan of the reservoir group.

[0011] In some embodiments, the IMOSSA is verified to be effective through a standard multi-objective test function, wherein the standard multi-objective test function includes but is not limited to Zitzler-Deb-Thiele (ZDT), Deb-Thiele-Laumanns-Zitzler (DTLZ), Schaffer, and Kursawe, and the number of objectives of the standard function is not less than 2.

[0012] In some embodiments, the method of using the entropy weight method to screen the global optimal solution, performing individual screening based on the reference point strategy, and using the binomial crossover method to update the finder individual, and determining an IMOSSA that solves the multi-objective optimization problem, includes the following steps:

[0013] Initialize the algorithm parameters, including determining the population size, the number of decision variables, the maximum number of iterations, the number of discoverers, the number of alerters, the population warning threshold and safety threshold; and initialize all individuals in the population;

[0014] Measure all individuals in the population, obtain the corresponding fitness values, and calculate the non-dominated sorting levels and crowding distances of all individuals; individuals with the same non-dominated sorting level are the same subset, and the non-dominated sorting set is obtained;

[0015] In the set with the largest non-dominated sorting level, the individual with the smallest crowding distance is selected as the current global worst solution; in the set with the smallest non-dominated sorting level, the current global optimal solution is obtained based on the entropy weight method;

[0016] Based on the improved sparrow algorithm, the current population is updated to obtain an updated population; the current population and the updated population are merged into the latest population, all individuals in the latest population are measured to obtain corresponding fitness values, the non-dominated sorting levels of all individuals are calculated, and the non-dominated sorting set is obtained;

[0017] According to the non-dominated sorting level, they are selected into the next generation of evolutionary populations in order from small to large, until the number of selected populations reaches the set population size, and the last selected non-dominated sorting level is marked as L 1; If the non-dominated sorting level L When all individuals with 1 are selected into the next generation evolutionary population, and the population size exceeds the set population size, the individual selection process is carried out according to the individual selection strategy based on the reference point to obtain the evolutionary population that will eventually enter the next generation;

[0018] Iterate the evolution until the number of iterations reaches the set maximum number of iterations, terminate the operation and output the proposed next generation evolution population.

[0019] In some embodiments, obtaining the current optimal solution based on the entropy weight method in the set with the smallest non-dominated sorting level includes the following steps:

[0020] Based on the number of individuals in the set with the smallest non-dominated sorting level and the evaluation index of each individual, a multi-attribute decision matrix is ​​constructed, and the decision matrix is ​​normalized based on a non-proportional transformation method to obtain a normalized matrix;

[0021] Determining an indicator entropy value based on the normalized matrix, and determining an entropy weight of each evaluation indicator based on the indicator entropy value;

[0022] Based on the normalized matrix and the entropy weight, determine and obtain the current optimal individual.

[0023] In some embodiments, updating the current population based on the improved sparrow algorithm includes the following steps:

[0024] Arrange the current population in positive order according to the non-dominated sorting level to define it as a new population; divide the new population into discoverers, followers and guards, where the discoverer is the front of the new population. individuals, and the number of discoverers is recorded as , followers are the remaining individuals in the new population, and guards are randomly selected individuals in the new population. N d individuals, and the number of vigilant individuals is recorded as N d ;

[0025] Combining the binomial mutation method to replace the standard sparrow algorithm update strategy, the following formula is used to update the discoverer:

[0026]

[0027] in,

[0028]

[0029] In the formula, g is the current iteration number, , Respectively expressed in g Daihe g +1 generation j Only the discoverer i The location of the dimension, and Individual i The upper and lower limits of the dimension, i= 1, 2, …, M , j =1, 2, ..., N e ; a is a random number between (0,1]; Q represents random numbers that follow a normal distribution, L Indicates 1× M The identity matrix of is the distribution index; They represent the warning threshold and safety threshold respectively;

[0030] The follower is updated using the following formula:

[0031]

[0032] In the formula, and Respectively represent the current global worst and best individuals in i dimensional position; A represents a 1x where all elements are randomly assigned 1 or -1 M Matrix of N is the total population;

[0033] Update the sentinel using the following formula:

[0034]

[0035] In the formula, β is a random number that follows a normal distribution; is a random number in the range of [-1, 1]; is the fitness value of the current j-th sentinel; Represents the current global worst individual fitness; ε is an infinitesimal constant to avoid the denominator being 0.

[0036] In some embodiments, the individual screening using a reference point-based strategy comprises the following steps:

[0037] The first L The objective function of the population individuals of level 1 is normalized and converted into a number between 0 and 1;

[0038] Defining reference points based on structured reference point generation method;

[0039] Calculate the L The distance between each individual in the population of level 1 and the reference point is used to realize the association operation between the individuals in the population and the reference point, and to calculate the number of microhabitats at each reference point;

[0040] According to the number of microhabitats at the reference point, L A certain number of individuals are selected from the population of level 1 to evolve into the next generation, so that the population size is exactly the set population size.

[0041] The second aspect of the present invention provides a reservoir group joint optimization dispatching system based on IMOSSA, which includes:

[0042] Algorithm optimization module, used to select the global optimal solution by entropy weight method, perform individual screening based on reference point strategy and update the discoverer individual by binomial crossover method, and determine an IMOSSAA that solves multi-objective optimization problems;

[0043] A data acquisition and preprocessing module is used to generalize the reservoir group system and determine the hydraulic and hydrological relationship between the reservoirs based on the collected and sorted basic information of the reservoir group system; wherein the basic information includes runoff data, the geographical location and basic characteristic parameters of the reservoir group and the river channel, the water level-storage capacity relationship curve, and the downstream water level-discharge flow relationship curve; and

[0044] A model building module is used to analyze the tasks of the reservoir group and establish a multi-objective joint optimization scheduling model for the reservoir group based on the hydraulic and hydrological relationship between the reservoirs; wherein the objective function of the multi-objective joint optimization scheduling model for the reservoir group includes at least two of the maximum power generation, the minimum water shortage, the minimum water abandonment, the maximum minimum output, the lowest operating water level in front of the reservoir, and the minimum reservoir discharge flow, and the constraint conditions include water balance, reservoir discharge flow, unit flow capacity, downstream flood control flow, reservoir water level, and non-negative constraints;

[0045] The execution module is used to execute the IMOSSA to solve the multi-objective joint optimization scheduling model of the reservoir group, take the water level value / flow of each reservoir in each time period as the decision variable, determine the upper and lower limits of the water level / flow value of each reservoir, divide the cycle time period of the overall scheduling of the reservoir group, obtain the Pareto non-inferior solution set, and determine the scheduling operation plan of the reservoir group.

[0046] A third aspect of the present invention provides an electronic device, wherein the electronic device includes a memory and a processor coupled to each other, and the processor is used to execute program instructions stored in the memory to implement the above-mentioned joint optimization scheduling method of a reservoir group.

[0047] A fourth aspect of the present invention provides a computer-readable storage medium having program instructions stored thereon, which, when executed by a processor, implement the above-mentioned method for joint optimization scheduling of a group of reservoirs.

[0048] The present invention adopts the above technical solution and has the following beneficial effects:

[0049] (1) Meeting the requirements of multi-objective joint optimization operation of reservoir groups;

[0050] (2) Using the reference point-based strategy to screen individuals, we can ensure that the non-inferior solution individuals are evenly distributed, especially when solving multi-dimensional objectives, with good diversity, thus accelerating global convergence;

[0051] (3) The entropy weight method is used to screen the global optimal solution, which can overcome the difficulty in selecting the optimal solution in the process of transforming from a single-objective to a multi-objective optimization algorithm;

[0052] (4) Using binomial crossover to replace the individual update method of the discoverer in sparrow search can prevent the population from falling into the local optimum and thus causing the algorithm to converge prematurely, thereby enhancing the global optimization performance of the algorithm;

[0053] (5) Before solving the optimization scheduling model, the algorithm is tested using standard functions to prove the robustness of the algorithm in solving multi-objective optimization problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 It is a flow chart of an embodiment of a reservoir group optimization scheduling method based on IMOSSA provided by the present invention;

[0056] Figure 2 It is a flow chart of another embodiment of the reservoir group optimization scheduling method based on IMOSSA provided by the present invention;

[0057] Figure 3 It is a flow chart of an embodiment of IMOSSA provided by the present invention;

[0058] Figure 4 This is the DTLZ1-4 function test result graph;

[0059] Figure 5 It is the spatial distribution map of the reservoir group joint dispatching scheme set;

[0060] Figure 6 It is a structural diagram of an embodiment of a scheduling system;

[0061] Figure 7 is a structural schematic diagram of an embodiment of an electronic device provided by the present invention;

[0062] Figure 8 It is a structural schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0063] The present invention is further explained below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to the present invention.

[0064] The present invention aims at the defects of traditional multi-objective quantum genetic algorithm that it is easy to fall into local convergence and individuals are difficult to converge to the frontier of non-inferior solutions, and provides a reservoir group joint optimization scheduling method based on IMOSSA (Improved multi-objective sparrow search algorithm). The method firstly uses the entropy weight method to screen the global optimal solution on the basis of the traditional single-objective troublesome search method, uses the binomial crossover method to update the individual of the discoverer, and uses the reference point strategy to screen the individual to complete the design of an improved multi-objective sparrow search method, and applies it to the multi-objective joint optimization scheduling of the reservoir group to obtain a variety of reservoir group scheduling schemes on the premise of good verification by the application of the multi-objective standard test function.

[0065] The present invention provides a reservoir group joint optimization scheduling method based on IMOSSA. In one embodiment, refer to Figure 1 , Figure 1 1 is a flow chart of an embodiment of a reservoir group optimization scheduling method based on IMOSSA provided by the present invention. The scheduling method comprises the following steps:

[0066] Step S01: Using the entropy weight method to screen the global optimal solution, using the reference point strategy to screen individuals and using the binomial crossover method to update the discoverer individuals, determine an IMOSSA that solves the multi-objective optimization problem;

[0067] Step S02: Based on the collected basic information of the reservoir group system, the reservoir group system is generalized to determine the hydraulic and hydrological relationship between the reservoirs;

[0068] Step S03: Based on the hydraulic and hydrological relationship between the reservoirs, analyze the tasks of the reservoir group and establish a multi-objective joint optimization scheduling model for the reservoir group;

[0069] Step S04: Execute the IMOSSA to solve the multi-objective joint optimization scheduling model of the reservoir group, take the water level value / flow of each reservoir in each time period as the decision variable, determine the upper and lower limits of the water level / flow value of each reservoir, divide the cycle time period of the overall scheduling of the reservoir group, obtain the Pareto non-inferior solution set, and determine the scheduling operation plan of the reservoir group.

[0070] In one embodiment, the IMOSSA determined in step S01 is verified by at least two standard multi-objective test functions to verify the validity of the IMOSSA. The standard multi-objective test functions include, but are not limited to, Zitzler-Deb-Thiele (ZDT) standard function, Deb-Thiele-Laumanns-Zitzler (DTLZ) standard function, Schaffer standard function, Kursawe standard function, etc. Further, the DTLZ standard function includes DTLZ1 standard function, DTLZ2 standard function, DTLZ3 standard function, DTLZ4 standard function, etc.

[0071] In another embodiment, if Figure 2 As shown, Figure 2 1 is a flow chart of an embodiment of a reservoir group optimization scheduling method based on IMOSSA provided by the present invention. The reservoir group joint optimization scheduling method based on IMOSSA comprises the following steps:

[0072] Step 1: Propose IMOSSA to solve multi-objective optimization problems and use a multi-objective test function to verify the effectiveness of the algorithm.

[0073] The improvement of IMOSSA lies in the use of entropy weight method to screen the global optimal solution, individual screening based on the reference point strategy, and binomial crossover method to update the discoverer individual.

[0074] like Figure 3 As shown, Figure 3 1 is a flow chart of an embodiment of IMOSSA provided by the present invention. IMOSSA solves multi-objective optimization problems and includes the following steps:

[0075] Step 1-1, initialize model parameters and determine population size , the number of decision variables M , maximum number of iterations MAXGEN , the number of discoverers , the number of alert individuals , the safety threshold of the population ST , distribution index ;

[0076] Step 1-2, let the number of iterations g =0, initialize each individual in the population to form the initial population ;

[0077] Step 1-3, measure the population All individuals in , get the corresponding fitness value, and calculate the non-dominated ranking of all individuals and crowding distance ; Individuals with the same non-dominated sorting level are the same subset, and the non-dominated sorting set is obtained ,in L is the current maximum non-dominated sorting level;

[0078] Step 1-4, the set with the largest non-dominated sorting level The individual with the smallest crowding distance is selected as the current worst solution. ; The set with the smallest level in non-dominated sorting F 1, the current optimal solution is obtained based on the entropy weight method , including the following steps:

[0079] Step 1-4-1 Establish a multi-attribute decision matrix. Assume F 1Inside m individuals, each of which includes n Evaluation indicators, constructing a decision matrix ,in i= 1, 2, …, n , j = 1, 2, ..., m ;

[0080] Step 1-4-2 Implement attribute normalization. Matrix based on non-proportional transformation method R Normalization is performed to ensure that each indicator attribute takes a value between 0 and 1, maintaining the importance ratio of the original indicator attribute value, and obtaining a normalized matrix ;

[0081] Step 1-4-3 Determine the index entropy value H i , the calculation formula is as follows:

[0082]

[0083] In the formula, .

[0084] Step 1-4-4 Calculate the entropy weight of each evaluation index , the calculation formula is as follows:

[0085]

[0086] Step 1-4-5 Multiply the decision matrix and the weight to obtain the solution evaluation result vector: ,in , select the individual corresponding to the maximum value as the optimal individual .

[0087] Step 1-5, based on the improved sparrow algorithm for the population Update and get new population ;merge and For new populations , measuring population All individuals in get the corresponding fitness value, calculate the non-dominated sorting level of all individuals, and obtain the non-dominated sorting set Among them, based on the improved sparrow algorithm Performing an update involves the following steps:

[0088] Step 1-5-1 Set the population Sorting by non-dominated sorting level is redefined as: population , dividing the population into discoverers P E , Followers P S and Vigilante P D , where the discoverer is the population forward N e individuals, and the followers are the remaining N - N e Individuals, the vigilant is the population Randomly selected from N d individual;

[0089] Step 1-5-2 Combined with the binomial mutation method, the following formula is used to update the discoverer:

[0090]

[0091] in,

[0092]

[0093] In the formula, g is the current iteration number, , Respectively expressed in g Daihe g +1 generation j Only the discoverer i The location of the dimension, and Individual i The upper and lower limits of the dimension, i= 1, 2, …, M , j =1, 2, ..., Ne ; a is a random number between (0,1]; Q represents random numbers that follow a normal distribution, L Indicates 1× M The identity matrix of is the distribution index. Represent the warning threshold value and safety threshold value respectively. R 2< ST When , it means that there are no other predators around, and the discoverer can update its position to make it easier to obtain food; when R 2≥ ST When , it means that the discoverer in the population has discovered other predators, and the entire population needs to fly to a safe area.

[0094] Step 1-5-3 Use the following formula to update the followers:

[0095]

[0096] In the formula, and Respectively represent the current global worst and best individuals in i dimensional position; A represents a 1× M The matrix of . N is the total population, when j > N / 2, indicating j The fitness value of a follower is low and it is far away from the optimal position, so it needs to fly to other places to find food; otherwise, the follower finds food near the optimal position.

[0097] Step 1-5-4 Use the following formula to update the sentinel:

[0098]

[0099] In the formula, β is a random number that follows a normal distribution; is a random number in the range of [-1, 1]; is the fitness value of the current j-th sentinel; Represents the current global worst individual fitness; ε is an infinitesimal constant to avoid the denominator being zero. quilt When in control, j A single vigilant individual is at the edge of the group and is vulnerable to attack by predators; otherwise, it starts to move closer to other individuals, thus reducing the risk of being attacked.

[0100] Step 1-6, select the dominant individuals in the front into the next generation evolutionary population according to the order of non-dominated sorting level In, until L The selected population size of the first layer reaches N , if you will L All individuals in the first layer are selected into the next generation evolutionary population, and the population size exceeds N , then the first step will be carried out according to the individual selection strategy based on the reference point. L The individual selection process at level 1 includes the following steps:

[0101] Step1-6-1 L The objective function of the population individuals in layer 1 is normalized and converted into a number between 0 and 1;

[0102] Step 1-6-2 Based on the structured reference point generation method, the reference points are evenly arranged in the normalized hyperplane. M dimensional target problem, if each target is divided into Z If the number of reference points is defined H Calculate according to the following formula;

[0103]

[0104] Step1-6-3 Before calculation L The distance between each individual in the first layer and the reference point is used to realize the association operation between the individuals in the population and the reference points, and to calculate the number of microhabitats at each reference point.

[0105] Step 1-6-4 Select the number of microhabitats from the reference point L In layer 1, a certain number of individuals are selected to evolve into the next generation population. , so that the population size is just N .

[0106] Step 1-7, determine the number of iterations g Has it been reached? MAXGEN If yes, terminate the operation and output the population If not, then g = g +1, return to Step 1-3 and continue evolving.

[0107] Step 2: Collect and organize basic information of the reservoir group system, generalize the reservoir group system, and determine the hydraulic and hydrological relationship between the reservoirs; the basic information includes: runoff data, geographical location and basic characteristic parameters of the reservoir group and river channel, water level-storage capacity relationship curve, downstream water level-discharge flow relationship curve;

[0108] Step 3: Analyze the tasks of the reservoir group and establish a multi-objective joint optimization scheduling model for the reservoir group based on the hydraulic and hydrological relationships between the reservoirs;

[0109]

[0110]

[0111] Among them, the objective function of the multi-objective joint optimization scheduling model of the reservoir group includes at least two of the maximum power generation, minimum water shortage, minimum water abandonment, maximum minimum output, lowest operating water level in front of the reservoir, and minimum reservoir discharge flow. The constraint conditions of the multi-objective joint optimization scheduling model of the reservoir group include water balance, reservoir discharge flow, unit flow capacity, downstream flood control flow, reservoir water level, and non-negative constraints.

[0112] In a specific embodiment, the tasks of a reservoir group, such as flood control and economic development, are analyzed, and a multi-objective joint optimization scheduling model for a reservoir group is established with maximum power generation and minimum output as objective functions, and with water balance, reservoir discharge flow, unit flow capacity, downstream flood control flow, reservoir water level, and non-negative constraints.

[0113] Step 4: Execute IMOSSA to solve the multi-objective joint optimization scheduling model of the reservoir group to obtain the Pareto non-inferior solution set and determine the scheduling operation plan of the reservoir group. The algorithm solves the reservoir scheduling model, selects the water level value / flow of each reservoir in each period as the decision variable, determines the upper and lower limits of the water level / flow value of each reservoir, and divides the cycle period of the overall scheduling of the reservoir group. T .

[0114] The joint optimization scheduling of the reservoir group of Reservoir A and Reservoir B is now taken as an example to illustrate the effectiveness and rationality of the invented method. Reservoir A is a multi-year regulation reservoir, with a dead water level of a1m, a normal water storage level of a2m, a storage capacity of a300 million m³, a flood limit water level of a4m, and a beneficial storage capacity of a500 million m³; Reservoir B is an annual regulation reservoir, with a dead water level of b1m, a normal water storage level of b2m, a flood limit water level of b3m, and a power station installed capacity of b40,000 kW. The two reservoirs undertake multiple tasks including power generation, water supply, and ice prevention. (Note, a1-a5 and b1-b4 are the specific parameter values ​​of Reservoir A and Reservoir B, respectively)

[0115] The present invention first selects DTLZ1-DTLZ4 standard functions for algorithm testing, wherein the DTLZ function information is shown in Table 1. The specific parameters of IMOSSA are set as: population size N =200, number of global iterations MAXGEN =10000, the number of decision variables is 30, the number of discoverers is 60, the number of alerters is 40, the population safety threshold is 0.8, and the distribution index is 20. Figure 4The figure shows the comparison between the Pareto and the actual Pareto (i.e., the Pareto corresponding to the standard function) after the IMOSSA algorithm runs 10 times, which are relatively close overall. At the same time, the HV and IGD indicators are used to evaluate the algorithm. The results are shown in Table 2. It can be seen that the IMOSSA algorithm has good robustness and is close to the actual Pareto. That is, the effectiveness of IMOSSA is verified by at least two standard multi-objective test functions.

[0116]

[0117] A joint optimization dispatching model for a reservoir group is further established with the goal of maximizing the total power generation of the reservoir group and the minimum power generation output, and with water balance, water level limit, generator flow capacity, ecological flow, ice prevention flow and other constraints; the discharge flow of each reservoir in each period is set as the decision variable, and IMOSSA is used to solve the optimization dispatching model. The parameters are the same as above, the number of decision variables is 24, and the spatial distribution of the dispatching scheme set is as follows Figure 5 The spatial distribution of the scheduling scheme set is a non-convex curve. The scheduling scheme is widely and evenly distributed. The total power generation of the reservoir group and the total minimum output of the reservoir group restrict and conflict with each other, and there is an obvious inverse relationship. Therefore, the optimal scheduling scheme of the reservoir group solved by the improved algorithm is reasonable and effective.

[0118] The present invention also provides a reservoir group joint optimization dispatching system based on IMOSSA, see Figure 6 , Figure 6 The scheduling system 100 includes an algorithm optimization module 10 , a data acquisition and preprocessing module 20 , a model building module 30 , and an execution module 40 .

[0119] The algorithm optimization module 10 is used to select the global optimal solution by using the entropy weight method, perform individual screening based on the reference point strategy, and use the binomial crossover method to update the individual finder, and determine an IMOSSAA that solves the multi-objective optimization problem. The data acquisition and preprocessing module 20 is used to generalize the reservoir group system based on the collected and sorted basic information of the reservoir group system, and determine the hydraulic and hydrological relationship between the reservoirs; wherein the basic information includes runoff data, the geographical location and basic characteristic parameters of the reservoir group and the river, the water level-storage capacity relationship curve, and the downstream water level-discharge flow relationship curve. The model building module 30 is used to analyze the tasks of the reservoir group and establish a multi-objective joint optimization scheduling model for the reservoir group based on the hydraulic and hydrological relationship between the reservoirs; wherein the objective function of the multi-objective joint optimization scheduling model for the reservoir group includes at least two of the maximum power generation, the minimum water shortage, the minimum water abandonment, the maximum minimum output, the lowest operating water level in front of the reservoir, and the minimum reservoir discharge flow, and the constraint conditions include water balance, reservoir discharge flow, unit flow capacity, downstream flood control flow, reservoir water level, and non-negative constraints. The execution module 40 is used to execute the IMOSSA to solve the multi-objective joint optimization scheduling model for the reservoir group, using the water level value / flow of each reservoir in each time period as the decision variable, determining the upper and lower limits of the water level / flow value of each reservoir, dividing the cycle period of the overall scheduling of the reservoir group, obtaining the Pareto non-inferior solution set, and determining the scheduling and operation plan of the reservoir group.

[0120] See also Figure 7 , Figure 7 1 is a schematic diagram of the structure of an embodiment of an electronic device provided by the present invention, wherein the electronic device 110 includes a memory 111 and a processor 112 coupled to each other, and the processor 112 is used to execute program instructions stored in the memory 112 to implement the reservoir group joint optimization scheduling method provided by the present invention. In a specific implementation scenario, the electronic device 110 may include but is not limited to: a display device, a microcomputer, and a server.

[0121] See also Figure 8 , Figure 8 It is a structural diagram of an embodiment of a computer-readable storage medium provided by the present invention. The computer-readable storage medium 120 stores program instructions 121, and when the program instructions 121 are executed by the processor, the reservoir group joint optimization scheduling method of the present invention is implemented.

[0122] In the several embodiments provided by the present invention, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on network units. Some or all of the units can be selected according to actual needs to achieve the purpose of this implementation scheme.

[0123] In addition, each functional unit in each embodiment of the present invention can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform all or part of the steps of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

[0124] The above is a detailed introduction to the reservoir group optimization scheduling method and system, equipment and storage medium based on IMOSSA provided by the embodiments of the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A reservoir group joint optimization scheduling method based on IMOSSA, characterized in that: The following steps are involved: The entropy weight method is used to screen the global optimal solution, the reference point strategy is used for individual screening, and the binomial crossover method is used to update the individual discoverer, and an improved multi-objective sparrow algorithm IMOSSA is determined to solve the multi-objective optimization problem. Based on the collected basic information of the reservoir group system, the reservoir group system is generalized to determine the hydraulic and hydrological relationship between the reservoirs; Based on the hydraulic and hydrological relationship between the reservoirs, the tasks of the reservoir group are analyzed and a multi-objective joint optimization scheduling model of the reservoir group is established; Execute the IMOSSA to solve the multi-objective joint optimization scheduling model of the reservoir group, take the water level value / flow of each reservoir in each time period as the decision variable, determine the upper and lower limits of the water level / flow value of each reservoir, divide the cycle period of the overall scheduling of the reservoir group, obtain the Pareto non-inferior solution set, and determine the scheduling operation plan of the reservoir group; The method of using the entropy weight method to screen the global optimal solution, using the reference point strategy to screen individuals, and using the binomial crossover method to update the finder individuals, determines an improved multi-objective sparrow algorithm IMOSSA for solving multi-objective optimization problems, including the following steps: Initialize the algorithm parameters, including determining the population size, the number of decision variables, the maximum number of iterations, the number of discoverers, the number of alerters, the population warning threshold and safety threshold; and initialize all individuals in the population; Measure all individuals in the population, obtain the corresponding fitness values, and calculate the non-dominated sorting levels and crowding distances of all individuals; individuals with the same non-dominated sorting level are the same subset, and the non-dominated sorting set is obtained; In the set with the largest non-dominated sorting level, the individual with the smallest crowding distance is selected as the current global worst solution; in the set with the smallest non-dominated sorting level, the current global optimal solution is obtained based on the entropy weight method; Based on the improved sparrow algorithm, the current population is updated to obtain an updated population; the current population and the updated population are merged into the latest population, all individuals in the latest population are measured to obtain corresponding fitness values, the non-dominated sorting levels of all individuals are calculated, and the non-dominated sorting set is obtained; According to the non-dominated sorting level, they are selected into the next generation of evolutionary populations in order from small to large, until the number of selected populations reaches the set population size, and the last selected non-dominated sorting level is marked as L 1; If the non-dominated sorting level L When all individuals with 1 are selected into the next generation evolutionary population, and the population size exceeds the set population size, the individual selection process is carried out according to the individual selection strategy based on the reference point to obtain the evolutionary population that will eventually enter the next generation; Iterate the evolution until the number of iterations reaches the set maximum number of iterations, terminate the operation and output the proposed next generation evolution population.

2. The reservoir group joint optimization scheduling method based on IMOSSA according to claim 1 is characterized in that: The IMOSSA is verified to be effective through a standard multi-objective test function, wherein the standard multi-objective test function includes at least two of the Zitzler-Deb-Thiele standard function, the Deb-Thiele-Laumanns-Zitzler standard function, the Schaffer standard function, and the Kursawe standard function.

3. The reservoir group joint optimization scheduling method based on IMOSSA according to claim 1 is characterized in that: The basic information of the reservoir group system includes runoff data, geographical location and basic characteristic parameters of the reservoir group and river channel, water level-storage capacity relationship curve, downstream water level-discharge flow relationship curve; and / or, The objective function of the multi-objective joint optimization scheduling model of the reservoir group includes at least two of the following: maximum power generation, minimum water shortage, minimum water abandonment, maximum minimum output, lowest operating water level in front of the reservoir, and minimum reservoir discharge flow. The constraints include water balance, reservoir discharge flow, unit flow capacity, downstream flood control flow, reservoir water level, and non-negative constraints.

4. The reservoir group joint optimization scheduling method based on IMOSSA according to claim 1 is characterized in that: The method of obtaining the current global optimal solution based on the entropy weight method in the set with the smallest non-dominated sorting level includes the following steps: Based on the number of individuals in the set with the smallest non-dominated sorting level and the evaluation index of each individual, a multi-attribute decision matrix is ​​constructed, and the decision matrix is ​​normalized based on a non-proportional transformation method to obtain a normalized matrix; Determining an indicator entropy value based on the normalized matrix, and determining an entropy weight of each evaluation indicator based on the indicator entropy value; Based on the normalized matrix and the entropy weight, determine and obtain the current optimal individual.

5. The reservoir group joint optimization scheduling method based on IMOSSA according to claim 1 is characterized in that: The updating of the current population based on the improved sparrow algorithm comprises the following steps: Arrange the current population in positive order according to the non-dominated sorting level to define it as a new population; divide the new population into discoverers, followers and guards, where the discoverer is the front of the new population. individuals, and the number of discoverers is recorded as , followers are the remaining individuals in the new population, and guards are randomly selected individuals in the new population. N d individuals, and the number of vigilant individuals is recorded as N d ; Combined with the binomial mutation method to replace the standard sparrow algorithm update strategy, the following formula is used to update the discoverer: , the follower is the remaining individuals in the new population, and the vigilant is randomly selected from the new population N d individuals, and the number of vigilant individuals is recorded as N d ; Combining the binomial mutation method to replace the standard sparrow algorithm update strategy, the following formula is used to update the discoverer: in, In the formula, g is the current iteration number, , Respectively expressed in g Daihe g +1 generation j Only the discoverer i The location of the dimension, and Individual i The upper and lower limits of the dimension, i= 1, 2, …, M , j =1, 2, ..., N e ; a is a random number between (0,1]; MAXGEN is the maximum number of iterations, M is the number of decision variables; is the distribution index; They represent the warning threshold and safety threshold respectively; The follower is updated using the following formula: In the formula, and Respectively represent the current global worst and best individuals in i dimensional position; A represents a 1x where all elements are randomly assigned 1 or -1 M Matrix of Q represents random numbers that follow a normal distribution; N is the total population; Update the sentinel using the following formula: In the formula, β is a random number that follows a normal distribution; is a random number in the range of [-1, 1]; is the fitness value of the current j-th sentinel; Represents the current global worst individual fitness; ε is an infinitesimal constant to avoid the denominator being 0.

6. The reservoir group joint optimization scheduling method based on IMOSSA according to claim 1 is characterized in that: The individual screening based on the reference point strategy comprises the following steps: The first L The objective function of the population individuals of level 1 is normalized and converted into a number between 0 and 1; Defining reference points based on structured reference point generation method; Calculate the L The distance between each individual in the population of level 1 and the reference point is used to realize the association operation between the individuals in the population and the reference point, and to calculate the number of microhabitats at each reference point; According to the number of microhabitats at the reference point, L A certain number of individuals are selected from the population of level 1 to evolve into the next generation, so that the population size is exactly the set population size.

7. A dispatching system based on the IMOSSA-based reservoir group joint optimization dispatching method according to any one of claims 1 to 6, characterized in that: include: The algorithm optimization module is used to select the global optimal solution using the entropy weight method, perform individual screening based on the reference point strategy, and use the binomial crossover method to update the discoverer individual, and determine an improved multi-objective sparrow algorithm IMOSSA to solve the multi-objective optimization problem; A data acquisition and preprocessing module is used to generalize the reservoir group system and determine the hydraulic and hydrological relationship between the reservoirs based on the collected and sorted basic information of the reservoir group system; wherein the basic information includes runoff data, the geographical location and basic characteristic parameters of the reservoir group and the river channel, the water level-storage capacity relationship curve, and the downstream water level-discharge flow relationship curve; and A model building module is used to analyze the tasks of the reservoir group and establish a multi-objective joint optimization scheduling model for the reservoir group based on the hydraulic and hydrological relationship between the reservoirs; wherein the objective function of the multi-objective joint optimization scheduling model for the reservoir group includes at least two of the maximum power generation, the minimum water shortage, the minimum water abandonment, the maximum minimum output, the lowest operating water level in front of the reservoir, and the minimum reservoir discharge flow, and the constraint conditions include water balance, reservoir discharge flow, unit flow capacity, downstream flood control flow, reservoir water level, and non-negative constraints; The execution module is used to execute the IMOSSA to solve the multi-objective joint optimization scheduling model of the reservoir group, take the water level value / flow of each reservoir in each time period as the decision variable, determine the upper and lower limits of the water level / flow value of each reservoir, divide the cycle time period of the overall scheduling of the reservoir group, obtain the Pareto non-inferior solution set, and determine the scheduling operation plan of the reservoir group.

8. An electronic device, characterized in that: The electronic device includes a memory and a processor coupled to each other, and the processor is used to execute program instructions stored in the memory to implement the reservoir group joint optimization scheduling method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: Program instructions are stored thereon, and when the program instructions are executed by a processor, the reservoir group joint optimization scheduling method described in any one of claims 1 to 6 is implemented.

Images