Medium and long term optimal scheduling method for integrated wind-solar-hydro storage system coupled with short term complementary operation

Abstract

The application belongs to the technical field of multi-purpose complementary optimal scheduling, and discloses a wind-solar-water-storage integrated medium-long term optimal scheduling method coupled with short-term complementary operation, which comprises the following steps: S1, common and multi-type objective functions of a wind-solar-water-storage integrated medium-long term optimal scheduling model are determined and integrated into the model; S2, decision variables of the model are determined, and a water balance calculation module of the medium-long term scale model, a water level adjustment module at the end of the scheduling period and a power output calculation module of the hydropower station are established; S3, an electric quantity balance module, a load matching coefficient calculation module and a new energy real consumption rate calculation module for coupling the short-term scale complementary operation model are established; the short-term scale model is used to calculate the real objective function fitness value from the short-term perspective with the solving results of the medium-long term model module as input conditions and boundaries, and then the solving results are returned to the medium-long term scale model; and S4, a heuristic algorithm is used for solving.

Description

Technical Field

[0001] The present invention belongs to the technical field of multi-energy complementary optimization scheduling, and relates to a medium- and long-term optimization scheduling model for wind, solar, and water storage. More specifically, it relates to a medium- and long-term optimization scheduling method for integrated wind, solar, and water storage coupled with short-term complementary operation. Background Art

[0002] Integrated optimal scheduling of wind, solar, hydro, and storage systems can ensure a stable power supply while improving energy efficiency and reducing unnecessary resource waste, such as curtailment. However, due to the high volatility and intermittent nature of wind and solar resources, and the complex spatiotemporal constraints involved in reservoir scheduling, the problem of joint optimal scheduling of wind, solar, hydro, and storage systems is extremely challenging. Traditional scheduling techniques and integrated optimization scheduling models (such as linear programming, dynamic programming, and their derivative algorithms) have significant limitations in their computational efficiency and solution capabilities when dealing with complex, multi-stage, and multi-dimensional integrated systems. This is particularly true when dealing with large-scale, multivariable, and constrained integrated energy systems, often leading to low computational efficiency and the "curse of dimensionality." Furthermore, conventional multi-energy complementary optimization scheduling models struggle to accurately capture the actual consumption of wind and solar power on a medium- to long-term scale. Therefore, developing an efficient, robust, and well-convergent medium- to long-term integrated optimization scheduling model for wind, solar, hydro, and storage systems that can meet the complex and changing scheduling requirements of multi-energy complementary systems is of great significance and practical value for improving the intelligent management of energy systems. Summary of the Invention

[0003] In response to the above-mentioned defects or improvement needs of the existing technology, the present invention provides a medium- and long-term optimization scheduling method for integrated wind, solar, water and storage systems with coupled short-term complementary operations. Its purpose is to solve the problems of medium- and long-term optimization scheduling of integrated wind, solar, water and storage systems, as well as the technical problems of poor model adaptability and difficulty in guiding the actual operation of the project.

[0004] To achieve the above objectives, according to one aspect of the present invention, a mid- to long-term optimization scheduling method for wind, solar, and hydropower storage systems coupled with short-term complementary operation is provided, comprising the following steps:

[0005] S1. Determine the objective function of the medium- and long-term optimization scheduling model for the integration of wind, solar, and hydropower. The optimization scheduling model for the multi-energy complementary system generally considers three types of objectives. One is the goal of focusing on the consumption of new energy, such as the consumption rate of wind and solar new energy, the amount of wind and solar new energy curtailment, the risk of wind and solar comprehensive curtailment, etc.; the second is the goal of focusing on the situation at the power transmission end, such as optimal load matching, residual load variance, smoothness of the power transmission process, etc.; the third is the goal of focusing on the power generation efficiency, such as maximizing the comprehensive power generation efficiency of hydropower, maximizing the power generation efficiency of wind and solar new energy, maximizing the system bundling transmission efficiency, etc. The above three types of objective functions have different collaborative or competitive relationships in different systems and different power generation and transmission situations. The model of the present invention is adaptive, and any number of objective functions of any type can be integrated into this model.

[0006] S2, determine the decision variable of the model as the discharge flow of the hydropower station, and use it to establish the water balance calculation module of the medium and long-term scale model, the water level adjustment module at the end of the scheduling period, and the hydropower station output calculation module.

[0007] S3 establishes a power balance module, a load matching coefficient calculation module, and a new energy real consumption rate calculation module for coupling the short-term complementary operation model. The short-term model uses the solution results of the medium- and long-term model module as input conditions and boundaries, calculates the actual objective function fitness value from a short-term perspective, and then returns it to the medium- and long-term model.

[0008] S4. The complexity of the medium- and long-term optimized scheduling method for wind, solar, and hydropower storage integrated with coupled short-term complementary operation proposed in the present invention depends on the dimension and number of decision variables. However, the complexity of solving the model of a general cascade hydropower system is very high due to its numerous constraints and huge number of decision variables, and a heuristic algorithm must be used to solve it.

[0009] Furthermore, the S2 specifically includes the constraint expressions of each module and the output calculation formula as follows:

[0010] (1), water balance constraint;

[0011] i,t = i,t-1 +(I i,t - i,t )·Δt

[0012] Where: V i,t-1 、V i,t are the water storage capacity of hydropower station i at the beginning and end of period t, m 3 ;I i,t is the inflow flow of hydropower station i in period t, m 3 / s;Qo i,t is the outflow of hydropower station i in period t, m3 / s; other symbols have the same meanings as before.

[0013] (2) Upper and lower limit constraints of the water level of the hydropower station;

[0014]

[0015] Where: Z i,t is the initial water level of hydropower station i in time period t, m; Z i,t is the lower limit of the initial allowable water level of hydropower station i in time period t, m; is the upper limit of the initial allowable water level of hydropower station i in time period t, m. Generally, the normal high water level is selected in the non-flood season, and the flood limit water level corresponding to flood control needs in the flood season; the meanings of other symbols are the same as before.

[0016] (3) Constraints on the initial and final water levels of the hydropower station;

[0017] Z i,1 =Z i,start

[0018]

[0019] Where: Z i,end 、 is the lower and upper limit of the water level at the end of the dispatching period of hydropower station i, m; Z i,start is the initial water level of hydropower station i during dispatching period; other symbols have the same meanings as before.

[0020] (4) Constraints on outflow from hydropower stations;

[0021]

[0022] Qo i,t =Qe i,t +Qs i,t

[0023] Where: Qo i,t and are the lower and upper limits of the discharge flow of hydropower station i in time period t, respectively, m 3 / s;Qe i,t and Qs i,t are respectively the power generation flow and abandoned water flow of hydropower station i in period t, m 3 / s; other symbols have the same meanings as before.

[0024] (5) Upper and lower limits of hydropower station output;

[0025]

[0026] Where: respectively, are the lower and upper limits of the allowed output of the hydropower station i at time period t, million kW, where the maximum output is generally the expected output of the hydropower station; other symbols have the same meanings as before.

[0027] (6) water level amplitude constraint of the hydropower station;

[0028]

[0029] wherein: is the upper limit of the water level amplitude, m, and the water level amplitude needs to be limited within a certain range due to the requirements of the environment of the hydropower station, shipping, etc.; other symbols have the same meanings as before.

[0030] (7) output calculation of the hydropower station;

[0031] P = KQH = 8.5 · Q release · ((H begin + H end ) / 2 - F H (Q release ))

[0032] wherein: P is the output of the hydropower station; K is the output coefficient, and by default, it is equal to 8.5; Q release is the discharge of the hydropower station at the time period, H begin and H end are the initial water level and the final water level of the hydropower station at the time period, respectively, F H (Q release ) is the tail water level of the hydropower station at the time period, which is a function affected by the discharge, and the value is obtained by interpolation of the discharge-tail water level curve.

[0033] Further, the S3 specifically comprises the constraint expression of each module of the model of the application and the output calculation formula as follows:

[0034] (1) power balance module

[0035] In the case of knowing the average hydropower generation of the time period at the medium and long term scale, the output process of the cascade hydropower station and the pumped storage power station is complementarily nonlinear programmed under the satisfaction of multiple constraint conditions such as the upper and lower limit constraints of the hydropower output, the climbing constraints of the hydropower unit, the efficiency constraints of the pumped storage power station, the capacity constraints of the pumped storage power station, etc., that is, the optimization of the load matching coefficient is realized by adjusting the time, size, etc. of the hydropower station output and the energy storage and power generation process of the pumped storage power station, and then the short-term hourly scale hydropower station output process and the energy storage and power generation process of the pumped storage power station are obtained, and the calculation of the real consumption rate of new energy and the calculation of the load matching process can be realized through the short-term correction process. The output of the hydropower station firstly needs to satisfy the constraint that the average value of the output in the short term is equal to the output at the medium and long term scale:

[0036]

[0037] Where: P i hydro =T0 is equal to the output of the hydropower station at time i, and T0 is equal to the length of the entire short-term correction time. For example, when the medium- and long-term scheduling time scale is 1 week, T0 is equal to 168 hours; Equal to the output of the hydropower station in the medium and long term.

[0038] (2) Wind power station output constraints

[0039]

[0040] Where: P t w is the average output of the wind power station in period t, kW; are the minimum and maximum outputs allowed for the wind power station in period t, in kW. The maximum output is generally the installed capacity of the power station.

[0041] (3) PV power station output constraints

[0042]

[0043] Where: are the minimum and maximum outputs allowed for the photovoltaic power station in period t, in kW. The maximum output is generally the installed capacity of the power station.

[0044] (4) Hydropower station output amplitude constraint

[0045]

[0046] Where: ΔP t h Indicates the hydropower output variation between adjacent time periods, kW; represents the upper limit of hydropower output variation, in kW. This constraint indicates that the hydropower output cannot respond in time due to limitations of the power generation equipment.

[0047] (5) Output constraints of pumped storage power stations

[0048]

[0049] In the formula, the left side Indicates the cumulative output of continuous energy storage or continuous power generation of the pumped storage power station. C It represents the installed capacity of the pumped-storage power station. The continuous energy storage or cumulative power generation output of the pumped-storage power station does not exceed the capacity constraint, and the energy for pumping water comes from the wind and solar power generation cluster connected to the pumped-storage power station.

[0050] (6) Load constraints at the transmission end

[0051] P t ≤D t

[0052] Where: D t is the grid load demand at time t, in kW. When the system output exceeds the grid load, the excess energy cannot be used, and power is abandoned.

[0053] According to another aspect of the present invention, a medium- and long-term optimization scheduling method for an integrated wind, solar, water and storage system with coupled short-term complementary operation is provided. When the program product is called and executed, a medium- and long-term optimization scheduling model and solution method for an integrated wind, solar, water and storage system with coupled short-term complementary operation as described in any of the previous items is implemented.

[0054] In general, the above technical solutions conceived by the present invention can achieve the following results compared with the prior art:

[0055] Beneficial effects:

[0056] The medium- and long-term optimization scheduling method for wind, solar, and hydropower storage integration with coupled short-term complementary operation proposed in the present invention can not only effectively improve the total bundled power transmission of the wind, solar, and hydropower storage integrated system and increase power generation efficiency when targeting multi-energy complementary optimization scheduling problems with high complexity, difficult constraints, and multi-objective combinations, but also exhibits excellent performance such as good convergence robustness, and produces a more reasonable complementary scheduling process while significantly increasing power generation. It is an effective and superior method product for solving multi-energy complementary optimization scheduling problems, and has important significance and practical value for improving the intelligent management level of energy systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 Schematic diagram of the process of the present invention.

[0058] Figure 2 This is a topological diagram of the wind, solar, water and storage integrated multi-energy complementary system in the embodiment.

[0059] Figure 3 Graph showing the final population and Pareto front distribution results after iteration in the embodiment.

[0060] Figure 4 This is a medium- and long-term scale hydropower output process diagram of the maximum individual solution of comprehensive cascade hydropower generation in the Pareto front in the embodiment.

[0061] Figure 5 This is a diagram of the operation process of cascade hydropower stations on a medium- and long-term scale, representing the maximum individual solution for comprehensive cascade hydropower generation in the Pareto front in the embodiment. DETAILED DESCRIPTION

[0062] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0063] like Figure 1 The flowchart of the medium- and long-term optimized scheduling method for wind, solar, and water storage integration with coupled short-term complementary operation is shown in the present invention, which includes the following steps:

[0064] S1. Determine the objective function of the medium- and long-term optimization scheduling model for the integration of wind, solar, and hydropower. The optimization scheduling model for the multi-energy complementary system generally considers three types of objectives. One is the goal of focusing on the consumption of new energy, such as the consumption rate of wind and solar new energy, the amount of wind and solar new energy curtailment, the risk of wind and solar comprehensive curtailment, etc.; the second is the goal of focusing on the situation at the power transmission end, such as optimal load matching, residual load variance, smoothness of the power transmission process, etc.; the third is the goal of focusing on the power generation efficiency, such as maximizing the comprehensive power generation efficiency of hydropower, maximizing the power generation efficiency of wind and solar new energy, maximizing the system bundling transmission efficiency, etc. The above three types of objective functions have different collaborative or competitive relationships in different systems and different power generation and transmission situations. The model of the present invention is adaptive, and any number of objective functions of any type can be integrated into this model.

[0065] S2, determine the decision variable of the model as the discharge flow of the hydropower station, and use it to establish the water balance calculation module of the medium and long-term scale model, the water level adjustment module at the end of the scheduling period, and the hydropower station output calculation module.

[0066] S3 establishes a power balance module, a load matching coefficient calculation module, and a new energy real consumption rate calculation module for coupling the short-term complementary operation model. The short-term model uses the solution results of the medium- and long-term model module as input conditions and boundaries, calculates the actual objective function fitness value from a short-term perspective, and then returns it to the medium- and long-term model.

[0067] S4, the complexity of the medium- and long-term optimization scheduling method for wind, solar, and hydropower storage integration with coupled short-term complementary operation proposed in the present invention depends on the dimension and number of decision variables. However, the complexity of model solution for general cascade hydropower systems is very high due to their numerous constraints and huge number of decision variables, and they must be solved using heuristic algorithms. Therefore, the algorithm program product used in the present invention for model solution is the NSGA-III algorithm, which combines decision variables of different time periods and different hydropower stations into individuals, simulates different feasible solutions by generating different individuals, and uses the reference point mechanism to determine the dominant relationship between individuals for multi-objective problems, while increasing the distribution uniformity of solutions to multi-objective problems. On this basis, crossover, mutation, and selection operators are used to realize the iteration of individuals in the optimization direction, and finally generate a set of multi-objective non-inferior solutions corresponding to different preference relationships, which reflects the mutual relationship and overall trend between the objectives as a whole.

[0068] In this embodiment, step S1 specifically includes:

[0069] In the calculation of this embodiment, the model adopts three objective functions, namely, maximizing the wind and solar energy new energy consumption rate, optimizing the load tracking effect, and maximizing the comprehensive power generation of cascade hydropower. The calculation formula is as follows:

[0070] (1) Wind and solar energy have the highest absorption rate

[0071] The wind and solar energy absorption rate (REUR) refers to the ratio of the actual bundled electricity delivered by wind and solar energy to the electricity generated by the power source at each time period. It can also be understood as the true rate at which wind and solar energy output is actually utilized. Because renewable energy output fluctuates and intermittently, grid-side load constraints and channel capacity constraints can lead to wind or solar power curtailment in some situations. The integrated, complementary, and optimized scheduling of renewable energy bases aims to maximize wind and solar power absorption through the storage and regulation of hydropower, reducing curtailment and, in other words, increasing the wind and solar energy absorption rate.

[0072]

[0073] In the formula, the first, second, and third terms of REUR represent the photovoltaic energy absorption rate, wind power energy absorption rate, and the absorption rate of the three photovoltaic clusters connected to the pumped storage power station, respectively; Indicates the expected output of photovoltaic energy, The difference between them is equal to the actual output of photovoltaic energy. Indicates the expected output of wind power energy, represents the curtailed power of wind power, and their difference is equal to the actual output of wind power; Indicates the electricity absorbed and stored by the pumped storage power station, represents the expected output of the photovoltaic cluster connected to the pumped storage power station; t represents a certain moment in the short-term correction process; T long-term Indicates the length of time for medium- and long-term scheduling. For example, when the medium- and long-term scheduling time scale is 1 week and the time span is 1 year, T long-term It is equal to 53 weeks; T short-term Indicates the length of time for short-term corrections. For example, when the medium- to long-term scheduling time scale is 1 week, T short-term =168 hours; all short-term and above output units nested within the medium- and long-term timescales are capacity units. It's important to note that while the model's purpose is to optimize the complementary scheduling of integrated wind, solar, and hydropower storage systems over the medium- and long-term timescales, calculating the true wind and solar power consumption rate requires considering the short-term timescale. This is because the volatility and intermittent nature of renewable energy output primarily manifests itself over the short-term. Simply averaging this output would lose its characteristics, making it impossible to calculate the true wind and solar power consumption rate.

[0074] (2) Load tracking effect is optimal

[0075] The load matching coefficient LMC can quantitatively evaluate the effect of load tracking. The calculation of LMC mainly examines the proportion of residual load:

[0076]

[0077] Where, It indicates the expected output of all energy sources in the channel during the period, and Load indicates the load value of the period.

[0078] (3) Cascade hydropower stations have the largest comprehensive power generation capacity

[0079]

[0080] Where: E hydro It is the comprehensive power generation of the hydropower station during the entire dispatching period. is the average comprehensive power generation of the ith hydropower station in the tth period, in units of electricity.

[0081] In this embodiment, step S2 specifically includes:

[0082] (1), water balance constraint;

[0083] i,t = i,t-1 +(I i,t - i,t )·Δt

[0084] Where: V i,t-1 、V i,t are the water storage capacity of hydropower station i at the beginning and end of time period t, m3 ;I i,t is the inflow flow of hydropower station i in period t, m 3 / s;Qo i,t is the outflow of hydropower station i in period t, m 3 / s; other symbols have the same meanings as before.

[0085] (2) Upper and lower limit constraints of the water level of the hydropower station

[0086]

[0087] Where: Z i,t is the initial water level of hydropower station i in time period t, m; Z i,t is the lower limit of the initial allowable water level of hydropower station i in time period t, m; is the upper limit of the initial allowable water level of hydropower station i in time period t, m. Generally, the normal high water level is selected in the non-flood season, and the flood limit water level corresponding to flood control needs in the flood season; the meanings of other symbols are the same as before.

[0088] (3) Constraints on the initial and final water levels of the hydropower station;

[0089] Z i,1 =Z i,start

[0090]

[0091] Where: Z i,end 、 is the lower and upper limit of the water level at the end of the dispatching period of hydropower station i, m; Z i,start is the initial water level of hydropower station i during dispatching period; other symbols have the same meanings as before.

[0092] (4) Constraints on outflow from hydropower stations;

[0093]

[0094] Qo i,t =Qe i,t +Qs i,t

[0095] Where: Qo i,t and are the lower and upper limits of the discharge flow of hydropower station i in time period t, respectively, m 3 / s;Qe i,t and Qs i,t are respectively the power generation flow and abandoned water flow of hydropower station i in period t, m 3 / s; other symbols have the same meanings as before.

[0096] (5) Upper and lower limits of hydropower station output;

[0097]

[0098] Where: are the lower and upper limits of the output allowed for hydropower station i in time period t, respectively, in 10,000 kW, where the maximum output is generally the expected output of the power station; the meanings of other symbols are the same as before.

[0099] (6) Constraints on the water level fluctuation of hydropower stations;

[0100]

[0101] Where: is the upper limit of water level fluctuation, m. Due to the requirements of hydropower station environment, shipping, etc., the water level fluctuation needs to be limited to a certain range; the meanings of other symbols are the same as before.

[0102] (7) Calculation of hydropower station output;

[0103] P=KQH=8.5·Q release ·((H begin +H end ) / 2-F H (Q release ))

[0104] Where: P is the output of the hydropower station; K is the output coefficient, which is equal to 8.5 by default; Q release is the discharge flow of the hydropower station during this period, H begin and H end are the initial and final water levels of the hydropower station during this period, F H (Q release ) is the tailwater level of the hydropower station during this period, which is a function affected by the discharge flow, and its value is obtained by interpolation of the discharge flow-tailwater level curve.

[0105] In this embodiment, step S3 specifically includes:

[0106] (1),Power balancing module;

[0107] When the average hydropower generation in a medium- to long-term period is known, and multiple constraints such as upper and lower limits of hydropower output, hydropower unit ramping constraints, pumped storage power station efficiency constraints, and pumped storage power station capacity constraints are met, a complementary nonlinear programming is performed on the output process of cascade hydropower stations and pumped storage power stations. That is, by adjusting the time and size of the hydropower station output and the energy storage and power generation output of the pumped storage power station, the optimal load matching coefficient is achieved, and then the short-term hourly scale hydropower station output process and the energy storage and power generation process of the pumped storage power station are obtained. Through the short-term correction process, the calculation of the true absorption rate of new energy and the calculation of the load matching process can be realized. The output of the hydropower station must first meet the constraint that the average output in the short term is equal to the output in the medium and long term:

[0108]

[0109] Where: P i hydro =T0 is equal to the output of the hydropower station at time i, and T0 is equal to the length of the entire short-term correction time. For example, when the medium- and long-term scheduling time scale is 1 week, T0 is equal to 168 hours; Equal to the output of the hydropower station in the medium and long term.

[0110] (2), wind power station output constraints;

[0111]

[0112] Where: P t w is the average output of the wind power station in period t, kW; are the minimum and maximum outputs allowed for the wind power station in period t, in kW. The maximum output is generally the installed capacity of the power station.

[0113] (3) ,PV power station output constraints;

[0114]

[0115] Where: are the minimum and maximum outputs allowed for the photovoltaic power station in period t, in kW. The maximum output is generally the installed capacity of the power station.

[0116] (4) hydropower station output amplitude constraint;

[0117]

[0118] Where: ΔP t h Indicates the hydropower output variation between adjacent time periods, kW; represents the upper limit of hydropower output variation, in kW. This constraint indicates that the hydropower output cannot respond in time due to limitations of the power generation equipment.

[0119] (5) Output constraints of pumped storage power stations;

[0120]

[0121] In the formula, the left side Indicates the cumulative output of continuous energy storage or continuous power generation of the pumped storage power station. C It represents the installed capacity of the pumped-storage power station. The continuous energy storage or cumulative power generation output of the pumped-storage power station does not exceed the capacity constraint, and the energy for pumping water comes from the wind and solar power generation cluster connected to the pumped-storage power station.

[0122] (6), load constraints at the transmission end;

[0123] P t ≤D t

[0124] Where: D t is the grid load demand at time t, in kW. When the system output exceeds the grid load, the excess energy cannot be used, and power is abandoned.

[0125] In this embodiment, step S4 specifically includes:

[0126] S41, Predefined Reference Points. The NSGA-III algorithm ensures solution diversity by predefined reference points. Reference points are distributed within the [0, 1] interval for each target. With "1" on each coordinate axis representing the maximum value of the corresponding target, [0, 1] represents the distribution range of target values. For each target, the [0, 1] interval is divided equally into a set number of segments to obtain the reference point distribution for that target. Reference points for different targets are combined to obtain reference points for the corresponding target spatial dimension. The predefined reference points serve as the basis for subsequent individual screening.

[0127] S42, initialization and generation of feasible solutions. According to the scheduling requirements of the cascade reservoir group, n individuals are generated to form the initial population, and each individual corresponds to a set of scheduling schemes. In the medium- and long-term scheduling process, the discharge flow of the hydropower station is selected as the decision variable. Then each individual in the population contains (T+1)×N decision variables, where T corresponds to the number of time periods in the scheduling period. Since the storage capacity at the end of the scheduling period needs to be considered, each power station corresponds to T+1 decision variables, and N corresponds to the number of hydropower stations that need to be regulated. The corresponding population matrix C pop As shown in the following formula, the population is used as the initial value of the iteration, that is, the first generation parent population.

[0128]

[0129] In the formula, ind represents the set of decision variables of individuals in the population, for example, for the nth individual ind n ; The decision variable set of each individual is a (T+1)×N matrix, It represents the discharge flow of the Nth hydropower station of the kth individual in the population at time T+1, in m 3 / s.

[0130] S43, crossover and mutation operators generate new individuals. The crossover operator randomly combines some of the decision variables of two individuals to form a new individual in order to find the local optimum. This paper uses the simulated binary crossover algorithm for calculation. The calculation methods are shown in the following formulas:

[0131]

[0132] Where: c l 、c u are parent population individuals l and u respectively; c l ' l and c u ' u are the new population individuals ll and uu generated by crossover; β is the recombination coefficient corresponding to the simulated binary crossover algorithm, and the randomness of its calculation corresponds to the random crossover process of genes.

[0133] The mutation operator changes the values ​​of some random variables within the feasible range to make them deviate from the original sequence in order to achieve the purpose of global random optimization. In this paper, the polynomial mutation algorithm is used for calculation. The calculation method is shown in the following formula:

[0134] c v ' v =c v +δ·(UL)

[0135] Where: δ is the polynomial coefficient of variation, corresponding to the random mutation process of the gene; U and L are the upper and lower limits of the decision variables respectively; c v is the parent population individual v; c v ' v The new population individual vv is generated by the mutation. After random crossover and mutation calculation, a new population of the same size n can be obtained. The parent population and the new population are merged to obtain an alternative population of size 2n.

[0136] After the above cross, mutation and merging, 2n alternative individuals can be obtained, but only n individuals can enter the next generation population, so the alternative population needs to be sorted and screened. In the NSGA-III algorithm, the screening is divided into two stages: first, according to the Pareto dominance principle, the non-dominated level is divided, and the non-inferior solution and the relatively good inferior solution are preferentially retained; for the remaining inferior solutions, the reference point method is further used for screening to ensure the uniformity of individual distribution.

[0137] S44, non-dominated sorting is carried out by using the Pareto dominance principle. t Through non-dominated sorting, the Pareto level of all solutions can be obtained, and according to the order from low to high of the Pareto level, the whole population of low level is preferentially put into the new parent population P t+1 , until the individuals of a certain level cannot be completely put into the new parent population P t+1 .

[0138] S45, further screening by using reference points. For the last level that cannot be completely put in S44, further screening of better individuals is still needed to put into the offspring population. The NSGA-III algorithm uses the reference point method for screening, which associates the individuals in the alternative population to the predefined reference points, and according to the density of individual distribution, the individuals in the relatively sparse place are preferentially selected and retained to ensure the uniformity of the offspring population distribution. It mainly includes the following steps:

[0139] First, normalize the target space: in order to eliminate the magnitude difference between different targets, the target value needs to be standardized first. The minimum value of each target is taken as the ideal point, and the target value of all individuals is subtracted from the ideal value to obtain the standard target value, that is, the target value is measured by the distance from the ideal point, as shown in the following formula.

[0140]

[0141] In the formula: f m (c l ) is the original target value of individual l target m; f m ′(c l ) is the standardized target value of individual l target m; is the minimum value of target m. On this basis, the scalarization function (ASF) is used to traverse the target value of all individuals to find the extreme point individual reflecting the maximum value of each target. Connecting each extreme point and extending it to the coordinate axis can obtain the intercept a m (the intersection of the hyperplane and the coordinate axis), which reflects the possible maximum value of each target, and the normalization operation is performed based on this, as shown in the following formula.

[0142]

[0143] Where: w m is the weight corresponding to the target m direction, which is used to obtain the extreme point closer to the target direction coordinate axis; f m ″(c l ) is the normalized target value for individual I and target m; f has the same meaning as above. Normalization converts individual target values ​​to the same calculation scale as the reference point, with each target intercept corresponding to the reference point boundary value of "1"; this is used for subsequent calculations.

[0144] Then associate the reference points and individual screening: According to the normalized individual target value, calculate the Euclidean distance of all individuals to the line connecting each reference point and the origin. The shortest distance indicates that the solution is associated with the reference point. In this way, all individuals can be assigned to different reference points, which is equivalent to partitioning the population. Count the number of individuals associated with each reference point ρ j , ρ j The larger the value, the denser the individuals are in the area, and vice versa. To ensure the globality and uniformity of the optimization results, the algorithm is based on ρ j Select reference points in order from small to large, and retain the individuals associated with them until the required individuals, and the individuals retained in steps (4) and (5) constitute the offspring population.

[0145] S46, convergence condition determination. Determine whether the target value difference between the obtained offspring population and the parent population is less than 1% or the predetermined number of iterations has been reached. If so, the optimization is complete. Otherwise, the selected offspring population is used as the new parent population, and the above steps are repeated until the above convergence conditions are met, thereby obtaining a set of mutually independent multi-energy complementary scheduling solutions.

[0146] Example: Take a wind, solar, and water storage integrated system as an example. Its system and power transmission structure are as follows: Figure 2 As shown. In order to compare and test the effectiveness and superiority of the present invention in the medium- and long-term optimization scheduling of multi-energy complementarity, this example designs two control groups, namely the conventional scheduling method using inflow and outflow balance in the same multi-energy complementary system (control group No. 1) and the hydropower separate optimization scheduling method (control group No. 2); the model calculation rule of control group No. 1 is to make the outflow of the hydropower station equal to the inflow, and the wind, solar, and hydropower complementary optimization is not performed when the wind, solar, and hydropower are transmitting electricity. The model calculation rule of control group No. 2 is to optimize the operation mode of the cascade hydropower station with the goal of maximizing the total power generation of cascade hydropower, and the wind, solar, and hydropower complementary optimization is not performed when transmitting electricity in the short term. The specific scheduling methods and solution information of the experimental group and the control group are shown in Table 1:

[0147] Table 1. Comparison of scheduling methods between the experimental group and the control group in the example design

[0148]

[0149] The scheduling period of the experimental group is set to one year, the scheduling time step of the medium- and long-term scale model is weekly (52 weeks in total), the scheduling time step of the short-term scale model is hourly (168 hours per week), the population size is set to 4000, the number of individuals for crossover and mutation is set to 2000, and the maximum number of iterations is 500. The inflow runoff of the cascade hydropower station is selected from the measured monthly runoff data of the dam site of Hydropower Station No. 1 in 2010. The photovoltaic and wind power outputs are both selected from the hourly scale simulated output data from 0:00 on January 1 to 23:00 on December 31, 2010. The load process uses a typical two-stage step curve. In this embodiment, the iteration termination condition in step S46 is to reach the maximum number of iterations. The basic information and constraints of the other relevant examples of the wind, solar and water storage integrated base are shown in Tables 2 to 4:

[0150] Table 2. Basic information of the wind, solar, and water storage integrated base

[0151]

[0152]

[0153] Table 3. Instance constraint set information table

[0154]

[0155] Table 4 shows the comparative analysis of the results between the experimental group and the control group. Figure 3A three-dimensional plot of the three-objective optimal scheduling calculated using an example of the present invention is presented. The dots represent a population of 4,000 individuals, and the stars represent the Pareto front distribution within that population. The chart shows that there are 61 individuals on the Pareto front, each with its own advantages and disadvantages. The optimal value of the maximum objective function for the renewable energy consumption rate (REUR) is 0.8689 (annual average, the same below), exceeding the REUR values ​​of the conventional scheduling solutions for control groups 1 and 2 by 5.83% and 9.31%, respectively. The optimal value of the maximum objective function for the load matching coefficient (LMC) is 0.8099, exceeding the LMC values ​​of the conventional scheduling solutions for control groups 1 and 2 by 8.34% and 4.42%, respectively. The optimal value of the maximum objective function for the combined cascade hydropower generation (Ehydropower) is 24.71646 billion kWh, exceeding the Ehydropower value of the conventional scheduling solution for control group 1 by 28.91%. Judging from the average results of the entire Pareto frontier, the average REUR value of all individuals in the frontier is equal to 0.8425, which is 2.61% and 5.98% higher than the REUR values ​​of the conventional scheduling solutions of the control groups No. 1 and No. 2, respectively; the average LMC value of all individuals in the frontier is equal to 0.7965, which is 6.55% and 2.69% higher than the LMC values ​​of the conventional scheduling solutions of the control groups No. 1 and No. 2, respectively; the average cascade hydropower comprehensive power generation Ehydropower of all individuals in the frontier is equal to 22.12748 billion kWh, which is 15.41% higher than the Ehydropower value of the conventional scheduling solution of the control group No. 1; the average annual cumulative bundled power output of the entire Pareto frontier solution set is equal to 49.53694 billion kWh, which is 5.84% and 2.18% higher than the control groups No. 1 and No. 2, respectively. Obviously, the medium- and long-term optimized scheduling model of wind, solar, and hydropower storage integrated with coupled short-term complementary operation proposed in the present invention can not only improve the new energy absorption rate, but also make the power transmission process more compatible with the load process, thereby increasing the cumulative bundled power output of the multi-energy complementary base throughout the year.

[0156] Table 4. Comparative analysis of the results between the experimental group and the control group

[0157]

[0158] From the perspective of the objective function results of individuals in the Pareto frontier, the optimal individual of the maximum objective function value of REUR, whose average LMC and comprehensive Ehydropower are equal to 0.7882 and 20.37526 billion kWh respectively, is 2.75% and 21.31% less than the corresponding values of the optimal individual of the maximum objective function of LMC and comprehensive Ehydropower respectively; the optimal individual of the maximum objective function value of LMC, whose average REUR and comprehensive Ehydropower are equal to 0.8092 and 245.3189 billion kWh respectively, is 7.38% and 0.75% less than the individual values of the maximum objective function respectively; the optimal individual of the maximum objective function value of comprehensive Ehydropower, whose average REUR and average LMC are equal to 0.8055 and 0.8067 respectively, is 7.88% and 0.39% less than the individual values of the maximum objective function respectively. It can be seen from the two-dimensional projection diagram that in the Pareto frontier, with the increase of the average REUR of individuals, the average LMC and comprehensive Ehydropower will decrease, although the degree of decrease and the rate of change are inconsistent, but it can reflect the overall competitive relationship between REUR and LMC, Ehydropower. And in the Pareto frontier, with the increase of the average LMC of individuals, the comprehensive Ehydropower shows an increasing trend as a whole, but when the LMC increases to a certain extent, the increasing trend of comprehensive Ehydropower slows down or even gradually decreases, which reflects the overall synergistic effect and local competitive relationship between LMC and Ehydropower. In general, the Pareto frontier solution set shows the synergistic and competitive relationship between the load matching, new energy consumption and the power generation benefit of water and electricity in the process of integrated scheduling of wind, light, water and storage, and in the specific scheduling process, the optimal solution can be selected according to the priority of different scheduling objectives. The above results show that the method proposed in the application can effectively improve the total power transmission of the integrated wind, light, water and storage system, increase the power generation benefit, and has strong stability, and the case analysis also fully proves the applicability and superiority of the application in the complementary optimization scheduling problem. In addition, the model of the application will also output the medium and long term operation process of the water and electricity station and the short term power output process of each power source in the optimal frontier, as shown in Figure 4 and Figure 5 The output results are saved in the form of Excel for decision makers to check.

[0159] In summary, when solving complex engineering problems such as wind, light, water and storage multi-energy complementary optimization scheduling, the application can exhibit good convergence robustness and other superior performance, and at the same time significantly increase the power generation while producing a more reasonable complementary scheduling process, which shows that the coupled short-term complementary operation of the wind, light, water and storage integrated medium and long term optimization scheduling method is an effective and superior method for solving multi-energy complementary optimization scheduling problems.

[0160] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A medium- and long-term optimization scheduling method for wind, solar, and hydropower storage systems coupled with short-term complementary operation, characterized by: The steps include: S1. Determine the objective function of the medium- and long-term optimization scheduling model for the integration of wind, solar, hydropower and storage. The optimization scheduling model of the multi-energy complementary system considers three types of objectives: one is the goal of focusing on the consumption of new energy; the second is the goal of focusing on the situation at the power transmission end; and the third is the goal of focusing on the power generation efficiency. S2, determine the model's decision variable as the hydropower station discharge, and use it to establish the water balance calculation module, the end-of-dispatching water level adjustment module, and the hydropower station output calculation module of the medium- and long-term scale model; S3: Establish a power balance module, a load matching coefficient calculation module, and a new energy real absorption rate calculation module for coupling the short-term complementary operation model; The role of the short-term scale model is to use the solution of the medium- and long-term model module as input conditions and boundaries, calculate the true fitness value of the objective function from a short-term perspective, and then return it to the medium- and long-term scale model; Specifically include: (1),Power balancing module; Given the average hydropower generation in the medium- and long-term time periods, and satisfying multiple constraints such as upper and lower limits of hydropower output, hydropower unit ramping constraints, pumped-storage power station efficiency constraints, and pumped-storage power station capacity constraints, a complementary nonlinear programming approach is implemented for the output processes of cascade hydropower stations and pumped-storage power stations. This involves adjusting the time and magnitude of the hydropower station output and the energy storage and power generation output of the pumped-storage power station to achieve the optimal load matching coefficient. This leads to the short-term hourly hydropower station output process and the energy storage and power generation process of the pumped-storage power station. The short-term correction process is then used to calculate the true absorption rate of new energy and the load matching process. The hydropower station output first satisfies the constraint that the average output in the short term is equal to the output in the medium- and long-term scales: Where: P i hydro = equal to the output of the hydropower station at time i, T0 equals the duration of the entire short-term correction; Equal to the hydropower station output on a medium to long-term scale; (2), wind power station output constraints; Where: P t w is the average output of the wind power station in period t, kW; are the minimum and maximum outputs allowed by the wind power station in period t, in kW; the maximum output is the installed capacity of the power station; (3) ,PV power station output constraints; Where: are the minimum and maximum outputs allowed for the photovoltaic power station in period t, in kW; the maximum output is the installed capacity of the power station; (4) hydropower station output amplitude constraint; Where: ΔP t h Indicates the hydropower output variation between adjacent time periods, kW; represents the upper limit of hydropower output variation, in kW. This constraint indicates that the hydropower output cannot respond in time due to the limitation of power generation equipment. (5) Output constraints of pumped storage power stations; In the formula, the left side Indicates the cumulative output of continuous energy storage or continuous power generation of the pumped storage power station. C Represents the installed capacity of a pumped-storage power station. The continuous storage or cumulative power generation output of the pumped-storage power station does not exceed the capacity constraint, and the energy for pumping water comes from the wind and solar power generation cluster connected to the pumped-storage power station. (6), load constraints at the transmission end; P t ≤D t Where: D t is the grid load demand at time t, in kW. When the system output exceeds the grid load, the excess energy cannot be used, and power is abandoned. S4, uses the NSGA-III algorithm to solve the medium- and long-term optimal scheduling model of wind, solar, and hydropower storage integration with coupled short-term complementary operation.

2. The mid- to long-term optimization scheduling method for wind, solar, and water storage integration coupled with short-term complementary operation according to claim 1 is characterized in that: The S2 specifically includes: (1), water balance constraint; V i，t =V i，t-1 +(I i，t -Yes i，t )·△t Where: V i,t-1 、V i,t are the water storage capacity of hydropower station i at the beginning and end of time period t, m 3 ;I i,t is the inflow flow of hydropower station i in period t, m 3 / s;Qo i,t is the outflow of hydropower station i in period t, m 3 / s; other symbols have the same meaning as before; (2) Upper and lower limit constraints of the water level of the hydropower station; Where: Z i,t is the initial water level of hydropower station i in time period t, m; Z i,t is the lower limit of the initial allowable water level of hydropower station i in time period t, m; is the upper limit of the initial allowable water level of hydropower station i in time period t, m, which is the normal high water level in the non-flood season and the flood limit water level corresponding to flood control needs in the flood season; (3) Constraints on the initial and final water levels of the hydropower station; WITH i,1 =Z i,start Where: Z i,end 、 is the lower and upper limit of the water level at the end of the dispatching period of hydropower station i, m; Z i,start is the initial water level of hydropower station i during dispatch period; (4) Constraints on outflow from hydropower stations; Qo i,t =What? i,t +Qs i,t Where: Qo i,t and are the lower and upper limits of the discharge flow of hydropower station i in time period t, respectively, m 3 / s;Qe i,t and Qs i,t are respectively the power generation flow and abandoned water flow of hydropower station i in period t, m 3 / s; (5) Upper and lower limits of hydropower station output; Where: are the lower and upper limits of the output allowed for hydropower station i in time period t, in 10,000 kW, where the maximum output is the expected output of the power station; (6) Constraints on the water level fluctuation of hydropower stations; Where: is the upper limit of water level fluctuation, m; (7) Calculation of hydropower station output; P=KQH=8.5·Q release ·((H begin +H end ) / 2-F H (Q release )) Where: P is the output of the hydropower station; K is the output coefficient, which is equal to 8.5; Q release is the discharge flow of the hydropower station during this period, H begin and H end are the initial and final water levels of the hydropower station during this period, F H (Q release ) is the tailwater level of the hydropower station during this period, which is a function affected by the discharge flow, and its value is obtained by interpolation of the discharge flow-tailwater level curve.

3. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method according to claim 1 or 2 are implemented.

4. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to claim 1 or 2 are implemented.

Images