Water-wind-light-storage collaborative short-term scheduling two-stage optimization method

Abstract

The invention belongs to the field of multi-energy complementary scheduling, and discloses a two-stage optimization method for water, wind, light and storage collaborative short-term scheduling. According to the main technical scheme, in the first stage, on the basis of generation of a water-wind-light combined scene, a water-wind-light combined short-term optimization scheduling model is constructed, and a day-ahead power generation plan of cascade hydropower and the power shortage or abandoning condition in each time period are given; in the second stage, the energy storage scale obtained in the first stage is used as an upper limit, the discrete series energy storage capacity is used as an input condition, a water-wind-solar-energy-storage joint optimization scheduling model is constructed, the energy storage scale needing to be configured by the power grid is further optimized, and then the effects of different energy storage capacities on reduction of power shortage and power abandoning of the power system are analyzed. Application verification is carried out by taking a Sichuan power grid water-wind-light-storage multi-energy complementary system as an example, and a result shows that accurate configuration of the energy storage capacity in different new energy scenes can be given, and decision support is provided for water-wind-light-storage collaborative planning and operation.

Description

Technical Field

[0001] The present invention belongs to the field of multi - energy complementary scheduling and relates to a two - stage optimization method for coordinated short - term scheduling of water, wind, light, and energy storage. Background Art

[0002] The integrated scheduling of water, wind, light, and energy storage has become an important research hotspot in the field of energy and power. The core lies in using flexible power sources such as hydropower and energy storage to balance the fluctuations of intermittent power generation such as wind energy and solar energy. By achieving the spatio - temporal complementary coordination of various types of power sources, the utilization rate of new energy is improved, the reliability of power supply is enhanced, and it strongly supports the "power supply guarantee and consumption promotion" of the new power system with high - penetration new energy. Therefore, how to configure energy storage and how the multi - energy complementary system operates have become problems that need to be solved urgently at present and for a long time to come.

[0003] At present, the research on the multi-energy complementary scheduling of water, wind, and light storage has received more attention at home and abroad, mainly focusing on the basin side or microgrids. On the basin side, around the water, wind, and clean energy bases in the basin, on the one hand, the mechanism of the complementary operation of water, wind, and clean energy in the base and the benefit changes among various power generation entities have been discussed (Jing Z, Wang Y, Chang J, et al. Benefit compensation of hydropower-wind-photovoltaic complementary operation in the large clean energy base [J]. Applied energy, 2024, 354: 122040.); on the other hand, various operation risks of the water, wind, and light base have been quantified, and multi-objective optimal scheduling of the base operation has been carried out based on risk indicators (Zhao Zhipeng, Yu Zhihui, Cheng Chuntian, et al. Multi-risk quantification and long-term multi-objective coordinated optimal scheduling method for water, wind, and light integrated base [J]. Automation of Electric Power Systems, 2024, 48(22): 118-130.). In terms of microgrids, most focus on the wind-solar-storage complementary system, where energy storage mainly refers to new energy storage. Scholars such as Nie et al. proposed a dynamic risk-limiting dispatching strategy based on demand response to optimize the economy, reliability, and environmental benefits of multi-energy microgrids (Nie Y, Qiu Y, Yang A, et al. Risk-limiting dispatching strategy considering demand response in multi-energy microgrids [J]. Applied energy, 2024, 353: 122088.); other studies take the economic benefits of microgrids as the goal, optimize the configuration of microgrid energy storage, and calculate the reliability of the microgrid capacity configuration (Li Yanzhe, Guo Xiaojia, Dong Haiying, et al. Capacity optimization configuration of hybrid energy storage system for wind / solar / storage microgrid [J]. Proceedings of the CSU-EPSA, 2020, 32(06): 123-128.). Most of the multi-type power supply systems involved in these research works have relatively small capacities, taking the deterministic new energy prediction results as the input. At the same time, research on large-scale water, wind, and light storage complementary systems for provincial power grids facing the multiple uncertainties of water, wind, and light is extremely rare, and the demand is very urgent. Summary of the Invention

[0004] To address the above problems, in order to reasonably evaluate the flexibility support capabilities of flexible power sources such as hydropower and energy storage in the power grid for uncertain new energy sources and the actual engineering configuration requirements, the present invention proposes a two-stage optimization method for coordinated short-term scheduling of hydropower, wind, and energy storage, and conducts application verification based on the multi-energy complementary system of the Sichuan power grid. The results show that the achievements of the present invention can reasonably plan the energy storage scale of the power grid, effectively transfer the peak power of new energy, and reduce the risks of power shortage and power curtailment in the power grid. Case analysis shows that under the scenario of the optimal energy storage scale, the daily average output deviation of the power grid is reduced by 41%.

[0005] The technical solution of the present invention:

[0006] A two-stage optimization method for coordinated short-term scheduling of hydropower, wind, and energy storage, comprising the following steps:

[0007] Step (1): Generate a combined hydropower, wind, and energy storage scenario according to the predicted and actual wind and solar power output and runoff data. The specific steps are as follows:

[0008] (1.1) Data preparation: Collect the predicted and actual data of wind and solar power output and runoff, calculate the prediction error, and statistically calculate the mean and covariance matrix of the prediction error for each time period.

[0009] (1.2) Error modeling and sampling: Based on the mean and covariance matrix of the error, use the multivariate normal distribution to generate random error samples.

[0010] (1.3) Scenario generation: Superimpose the prediction error on the random error sample to generate a combined hydropower, wind, and energy storage scenario.

[0011] Step (2): Establish a short-term combined hydropower, wind, and energy storage scheduling model as the stage-one model. The model is as follows:

[0012] Objective function:

[0013]

[0014] In the formula: ns is the number of scenarios; s is the scenario number; t is the time period number; short t,s is the power shortage output of scenario s at time period t, MW; curtail t,s is the power curtailment output of scenario s at time period t, MW.

[0015] The constraint conditions include hydraulic constraints and power constraints. Among them, the hydraulic constraints include water volume balance constraints, water level control constraints at the beginning and end of the scheduling period, upper and lower water level limits, water level-storage relationship, upper and lower limits of power generation flow, upper limit of outflow flow, and upper and lower limits of power station output. The power constraints include power balance constraints and transmission line constraints.

[0016] Hydropower constraints:

[0017]

[0018] Where: V m,t+1 and V m,t are the water storage volumes (reservoir capacities) of hydropower station m at the beginning of time periods t + 1 and t, respectively, in m 3 ; Q m,t and S m,t are the inflow and outflow discharges of reservoir m during time period t, respectively, in m 3 / s; Δt is the time step of the calculation period. Z m,0 represents the initial water level of the dispatching period of hydropower station m, in m. Z m,T represents the terminal water level of the dispatching period of hydropower station m, in m. Z m,t and represent the lower and upper limits of the water level Z m,t of hydropower station m during time period t, respectively. In the non-flood season, it is taken as the dead water level to the normal high water level, and in the flood season, it is taken as the dead water level to the flood limit water level. f represents the relationship between the water level and the reservoir capacity of reservoir m. q m,t and represent the lower and upper limits of the power generation discharge q m,t of hydropower station m during time period t, respectively, in m 3 / s. S m,t and represent the lower and upper limits of the outflow discharge S m,t of hydropower station m during time period t, respectively, in m 3 / s. and represent the lower and upper limits of the power output of hydropower station m during time period t, respectively, in MW.

[0019] Power constraint:

[0020]

[0021] Where: Load is the total load of various power sources in the dispatching area, including the power load sent out, in MW; is the power output of all hydropower stations at time t in scenario s, in MW; is the power output of all wind power stations at time t in scenario s, in MW; is the power output of all photovoltaic power stations at time t in scenario s, in MW; P l,t,s is the channel power output of line l in scenario s during time period t, in MW; is the power output of hydropower station m at time t in scenario s; M is the number of hydropower stations in the calculation area; M l is the number of hydropower stations in channel l within the calculation area; is the sum of the power outputs of other power sources except hydropower for line l in scenario s during time period t; P l,max is the maximum transmission capacity of channel l. Different power stations in cascade hydropower may transmit power outward through different lines.

[0022] Step (3) takes the combined water-wind-solar scenario generated in step (1) as the input condition, solves the stage-one model in (2), and obtains the input parameters of the stage-two model. The specific steps are as follows:

[0023] (3.1) Considering the current and future scenarios of the installed capacity of the two new energy sources, generate the combined water-wind-solar scenario through the method in step (1), and use the generated scenario as the input condition of the stage-one model. Use the commercial solver gurobi to solve the stage-one model to obtain the output process of each power source and the load tracking situation (discarded power / power shortage per time period).

[0024] (3.2) Based on the load tracking situation in the future wind-solar installed capacity scenario calculated in step (3.1), evaluate the discarded power and power shortage in this scenario. Combine the discarded power and power shortage (take the larger of the two) to obtain the energy storage scale required to achieve complete power transfer, and use this energy storage scale as the upper limit of the energy storage capacity in the second stage.

[0025] Step (4) establishes a combined water-wind-solar-energy storage dispatch model as the stage-two model. The objective function, hydropower constraints, and power constraints are the same as those of the stage-one model. The energy storage constraints are as follows:

[0026]

[0027]

[0028] In the formula: r 0 represents the energy storage charge state variable, which is a 0-1 variable; r 1 represents the energy storage discharge state variable, which is a 0-1 variable. represents the output of the energy storage at time t in scenario s, MW; represents the power generation power of the energy storage at time t in scenario s, MW; represents the charging power of the energy storage at time t in scenario s; represents the maximum discharge power of the energy storage; represents the maximum charging power of the energy storage. In this calculation, the reduction of the energy storage discharge power limit with the decrease of the battery power is not considered.

[0029] The energy storage utilization rate index is used to evaluate the utilization of the energy storage and analyze the rationality of configuring different scales of energy storage. The index calculation formula is shown in formula (6).

[0030]

[0031] In the formula: UR is the energy storage utilization rate; represents the average value of the energy storage charge / discharge power, MW; P max represents the maximum value of the configured energy storage charge / discharge power.

[0032] Step (5) Based on the upper limit of the energy storage scale solved in Phase I of Step (3), different scales of energy storage are given as input conditions to solve the model in Phase II, and the energy storage utilization rate index is calculated. The specific steps are as follows:

[0033] (5.1) Consider two energy storage scale scenarios: one is to configure energy storage according to the upper limit of power output for power shortage or curtailment (the larger of the two), and the other is to configure energy storage according to the average power output for curtailment and the average power output for power shortage.

[0034] (5.2) Use the commercial solver gurobi to solve the model in Phase II to obtain the power output processes of each power source, the charge-discharge processes of the energy storage, and the load tracking situation (power curtailment / power shortage by period).

[0035] (5.3) According to the charge-discharge process of the energy storage, calculate the energy storage utilization rate UR according to Equation (6). By comparing the effects of different scales of energy storage on reducing power shortage and curtailment in the power grid, determine the construction plan for the energy storage supporting the power grid side in the future.

[0036] Advantages of the present invention:

[0037] The present invention provides a scientific method to optimize the short-term scheduling of hydropower, wind power, photovoltaic and energy storage resources through a two-stage scheduling model, thereby improving the flexibility and operation efficiency of the system. In the first stage, the present invention considers the existing situation, conducts short-term optimal scheduling for a power system with hydropower as the main flexible power source, and evaluates the possible problems existing in the current stage of the power system; in the second stage of the invention, considering the future development trend, it provides guidance for the energy storage configuration in the scenario of large-scale grid connection of wind and light in the future. Description of the Drawings

[0038] Figure 1(a) to Figure 1(d) is the combined scheduling result diagram of hydropower, wind power and photovoltaic, in which Fig. 1(a) is the load tracking diagram in the dry season, Fig. 1(b) is the load tracking diagram in the flood season, Fig. 1(c) is the output deviation diagram in the dry season, and Fig. 1(d) is the output deviation diagram in the flood season;

[0039] Figure 2(a) to Figure 2(b) is the combined scheduling result diagram of 2.5 times photovoltaic hydropower, wind power and photovoltaic, in which Fig. 2(a) is the load tracking diagram in the 2.5 times photovoltaic scenario, and Fig. 2(b) is the output deviation diagram in the 2.5 times photovoltaic scenario;

[0040] Figure 3(a) to Figure 3(d) is the combined scheduling diagram of energy storage configuration, in which Fig. 3(a) is the load tracking diagram of 600MW energy storage configuration, Fig. 3(b) is the load tracking diagram of 2200MW energy storage configuration, Fig. 3(c) is the change curve of energy storage utilization rate, and Fig. 3(d) is the output deviation diagram in the scenario of configured energy storage. Detailed Implementation Manner

[0041] The following further illustrates the detailed implementation manner of the present invention in combination with the drawings and technical solutions.

[0042] A two-stage optimization method for short-term coordinated scheduling of water, wind, and light energy storage, comprising the following steps:

[0043] (1) Generate a combined water, wind, and light energy scenario based on predicted and actual wind and light energy output and runoff data. The specific steps are as follows:

[0044] Step1: Data preparation: Collect predicted and actual data on wind and light energy output and runoff, calculate the prediction error, and statistically calculate the mean and covariance matrix of the prediction error for each time period.

[0045] Step2: Error modeling and sampling: Based on the mean and covariance matrix of the error, generate random error samples using the multivariate normal distribution.

[0046] Step3: Scenario generation: Superimpose the prediction error on the random error samples to generate a combined water, wind, and light energy scenario.

[0047] (2) Establish a short-term combined scheduling model for water, wind, and light energy as the first-stage model. The model is as follows:

[0048] Objective function:

[0049]

[0050] In the formula: ns is the number of scenarios; s is the scenario number; t is the time period number; short t,s is the power shortage output of scenario s at time period t, MW; curtail t,s is the abandoned power output of scenario s at time period t, MW.

[0051] The hydropower constraints specifically include:

[0052] Water balance constraint:

[0053] V m,t+1 =V m,t +3600(Q m,t -S m,t )Δt; t∈[1,T], m∈[1,M] (8)

[0054] In the formula: V m,t+1 and V m,t are the water storage volumes at the beginning of time periods t+1 and t for hydropower station m, respectively, m 3 ; Q m,t and S m,t are the inflow and outflow discharges of reservoir m at time period t, respectively, m 3 / s, where Δt is the time step of the calculation period. The inflow of reservoir m at time t consists of two parts: the regional flow of reservoir m and the outflow of the directly upstream reservoir m - 1 considering the water flow delay. The present invention does not consider the water flow delay; T is the number of calculation periods; M is the number of hydropower stations.

[0055] Initial water level control constraint during the scheduling period:

[0056]

[0057] In the formula: Z m,0 represents the starting water level of hydropower station m during the scheduling period, m.

[0058] Final water level control constraint during the scheduling period:

[0059]

[0060] In the formula: Z m,T represents the ending water level of hydropower station m during the scheduling period, m.

[0061] Reservoir water level constraint:

[0062]

[0063] In the formula: Z m,t and respectively represent the lower limit and upper limit of the water level Z m,t of hydropower station m at time t. During the non - flood season, it is taken as the dead water level to the normal high water level, and during the flood season, it is taken as the dead water level to the flood limit water level.

[0064] Water level - storage capacity constraint:

[0065] V m,t = f(Z m,t ); t ∈ [1, T], m ∈ [1, M] (12)

[0066] In the formula: V m,t represents the storage capacity of hydropower station m at time t, m 3 . f represents the water level - storage capacity relationship of reservoir m.

[0067] Power generation flow constraint:

[0068]

[0069] In the formula: q m,t and respectively represent the lower limit and upper limit of the power generation flow q m,t of hydropower station m at time t, m 3 / s.

[0070] Outflow constraint:

[0071]

[0072] Where: S m,t and respectively represent the lower limit and upper limit of the outflow discharge S of hydropower station m in period t, m m,t / s. 3 / s.

[0073] Constraints on the upper and lower limits of power station output:

[0074]

[0075] Where: and respectively represent the output of hydropower station m in period t lower limit and upper limit, MW.

[0076] Power constraints specifically include:

[0077] Power balance constraint:

[0078]

[0079] Where: Load is the total load of various power sources in the dispatching area, including the externally transmitted load, MW; The output of all hydropower stations at time t in scenario s, MW; The output of all wind power stations at time t in scenario s, MW; The output of all photovoltaic power stations at time t in scenario s, MW;

[0080] Transmission line constraint:

[0081]

[0082] Where: P l,t,s is the channel output of line l in scenario s at period t, MW; is the output of hydropower station m at time t in scenario s; M is the number of hydropower stations in the calculation area; M l is the number of hydropower stations in channel l within the calculation area; is the sum of the outputs of other power sources except hydropower for line l in scenario s at period t; P l,max is the maximum power transmission capacity of channel l. Different power stations in cascade hydropower may transmit power outward through different lines.

[0083] (3) Take the water-wind-solar scenario generated in step (1) as the input condition, solve the stage-one model in step (2), and obtain the input parameters for stage two. The specific steps are as follows:

[0084] Step 1: Consider two scenarios of current and future new energy installed capacity. Generate the combined scenario of hydropower, wind power, and photovoltaic power through the method in step (1), and use the generated scenario as the input condition of the stage 1 model. Solve the stage 1 model using the commercial solver Gurobi to obtain the output process of each power source and the load tracking situation.

[0085] Step 2: Based on the load tracking situation under the future installed capacity scenario of wind and photovoltaic power calculated in Step 1, evaluate the curtailed power and the power shortage of this scenario. Combine the curtailed power and the power shortage (take the larger of the two) to obtain the energy storage scale required to achieve complete power transfer, and use this energy storage scale as the upper limit of the energy storage capacity in stage 2.

[0086] (4) Establish a combined dispatching model of hydropower, wind power, photovoltaic power, and energy storage as the stage 2 model. The objective function, hydropower constraints, and power constraints are the same as those in the stage 1 model. The energy storage constraints are as follows:

[0087] Energy storage state constraint:

[0088]

[0089] In the formula, r 0 represents the energy storage charging state variable, which is a 0-1 variable; r 1 represents the energy storage discharging state variable, which is a 0-1 variable.

[0090] Energy storage output constraint:

[0091]

[0092] In the formula, represents the output of the energy storage at time t in scenario s, MW; represents the power generation power of the energy storage at time t in scenario s, MW; represents the charging power of the energy storage at time t in scenario s; represents the maximum discharging power of the energy storage; represents the maximum discharging power of the energy storage. In this calculation, the reduction of the energy storage discharging power upper limit with the decrease of the battery power is not considered.

[0093] Power balance constraint:

[0094]

[0095] Adopt the energy storage utilization rate index to evaluate the utilization of the energy storage, and analyze the rationality of configuring different scales of energy storage. The calculation formula of the index is shown in formula (21).

[0096] Energy storage usage index:

[0097]

[0098] Where: UR is the energy storage utilization rate; represents the average value of the energy storage charge / discharge power, MW; P max represents the maximum value of the configured energy storage charge / discharge power.

[0099] (5) According to the upper limit of the energy storage scale solved in the first stage, given different scales of energy storage as input conditions, solve the model in the second stage, and calculate the energy storage utilization rate index. The specific steps are as follows:

[0100] Step1: Consider two energy storage scale scenarios: one is to configure energy storage according to the upper limit of the power output of power shortage or curtailment (take the larger of the two), and the other is to configure energy storage according to the average power output of curtailment and the average power output of power shortage.

[0101] Step2: Use the commercial solver gurobi to solve the model in the second stage to obtain the power output process of each power source, the energy storage charge / discharge process, and the load tracking situation (power curtailment / power shortage by time period).

[0102] Step3: According to the energy storage charge / discharge process, calculate the energy storage utilization rate UR according to Equation (21). By comparing the effects of different scales of energy storage on reducing power shortage and curtailment in the power grid, determine the construction plan of the energy storage supporting the power grid side in the future.

[0103] This embodiment uses the provincial dispatching power sources of the Sichuan Power Grid to verify the feasibility of the model. The configuration principles of energy storage facilities are also considered to adapt to the expansion of future wind and solar power generation devices. The installed capacity of various power sources is as follows: wind power 7700MW, photovoltaic power generation 5400MW, and hydropower 14100MW.

[0104] Joint dispatching results of hydropower, wind power and photovoltaic power (Stage 1)

[0105] By substituting the typical days in the dry season and the typical days in the flood season of the research area into the model for solution respectively, the Figure 1(a) to Figure 1(d) joint dispatching results of hydropower, wind power and photovoltaic power are obtained. Among them, Figure 1(a) is the dispatching result in the dry season, and Figure 1(b) is the dispatching result in the flood season. Then, according to the actual power output values of wind and light, the power output deviation between the actual power output and the load on the same day is calculated to obtain Figure 1(c) and Figure 1(d).

[0106] As can be seen from Figure 1(a), there is a large gap between the load tracking curve (black line) and the combined output (superimposed area) of hydropower, wind power and photovoltaic power in the dry season. The average power shortage amount in each time period within a day is 2000MW. The main reason for the power shortage is the insufficient inflow of the basin, resulting in the fact that the hydropower output cannot fully meet the load demand. The average power shortage amount of the wind and light power plants under the large scenario (high wind and light power output) is 2520MW, while the average power shortage amount under the small scenario (low wind and light power output) is relatively high, at 1407MW. This shows that the fluctuation of wind and light resources has a greater impact on power shortage.

[0107] As can be seen from Figure 1(b), the tracking effect of hydropower load during the flood season is good. It can be seen from the figure that the combined output (area superposition) can almost completely cover the load curve (black line), and the power shortage problem is significantly reduced. The average curtailed power of the wind-solar power plant in the large scenario is 618 MW, and the average power shortage in the small scenario is 654 MW. As for the extremely small output deviation from the actual output scenario, it is because the generated wind-solar scenario cannot accurately predict the actual wind-solar output of the next day. Therefore, improving the prediction accuracy is particularly important for improving the accuracy of the day-ahead plan.

[0108] The power shortage problem during the dry season is significant, mainly due to insufficient hydropower resources and large fluctuations in wind-solar output. Especially in the small wind-solar scenario, the power shortage phenomenon is more serious. During the flood season, due to the sufficiency of hydropower resources, the load tracking effect is good, and the power shortage is greatly reduced. The consumption capacity of wind-solar resources has been improved, but there is still a certain amount of curtailed power in the large scenario. Therefore, it is necessary to further improve the prediction accuracy of wind-solar resources.

[0109] Considering the power shortage situation caused by insufficient water inflow during the dry season and the future energy structure of the region, large-scale grid connection of photovoltaic power plants will occur in the future. Due to the intra-day volatility of photovoltaic power, it will bring greater challenges to flexible power sources. Through calculation, when the regional photovoltaic installed capacity increases to 2.5 times the original, it can meet the power demand of a typical day during the dry season. The dispatching results are shown in Figures 2(a) and 2(b).

[0110] Due to the large peak-valley difference in the intra-day output of photovoltaic power, when the installed capacity of photovoltaic power in the system increases, curtailment is likely to occur at noon, and there is still an increase in power shortage in the morning and evening. When the photovoltaic is increased to 2.5 times the installed capacity, the average curtailed power from 9:00 to 16:00 is 990 MW, and the average power shortage in the remaining periods is 535 MW. The total power shortage for the whole day is 8572 MWh. Based on this, considering the application of the water-wind-solar-storage combined dispatching model in the second stage to solve the impact of the addition of energy storage with different performances in the power grid on grid connection.

[0111] Water-wind-solar-storage combined dispatching results (second stage)

[0112] In the second stage, two energy storage devices with different performances and a duration of 5 hours are configured to compare the differences in their effects during use.

[0113] As can be seen from the output curves (Figs. 3(a) and 3(b)), different energy storage capacities have a significant effect on improving the matching between system output and load. Without energy storage configuration, the volatility of wind and solar power generation is large, and there are obvious deviations between the output curve and the load curve, resulting in a high system power shortage and curtailment. After configuring 600 MW of energy storage, the energy storage system discharges during peak hours and charges during low valley hours, effectively realizing the time transfer of electricity, significantly reducing the deviation of the output curve, and reducing the daily average output deviation by 41%. When the energy storage capacity is further increased to 2200 MW, the output curve almost completely fits the load curve, and the deviation is basically eliminated. Only a small amount of power shortage mainly comes from the prediction error of wind and solar power generation.

[0114] Fig. 3(c) shows the discharge and charge utilization rates of the energy storage system under different energy storage configurations, further revealing the impact of energy storage capacity on system efficiency. Specifically, when 600 MW of energy storage is configured, both the discharge and charge utilization rates are 76%, indicating that the energy storage system can be efficiently utilized at this capacity. However, when the energy storage capacity increases to 2200 MW, the discharge utilization rate drops to 59%, and the charge utilization rate decreases to 37%, indicating that the utilization efficiency of the energy storage system decreases under high-capacity configurations, and there is a certain degree of underutilization.

[0115] Based on the above analysis, in the planning and construction of energy storage systems, in addition to improving system dispatching stability, the balance between energy storage capacity and utilization efficiency, as well as the construction cost of energy storage systems, should be fully considered to achieve higher economic efficiency and resource utilization efficiency.

Claims

1. A two-stage optimization method for short-term coordinated scheduling of water, wind, solar and storage, characterized in that: The steps include: Step (1) Generate a water-wind-solar joint scenario based on the predicted and actual wind-solar output and runoff data. The specific steps are as follows: (1.1) Data preparation: Collect forecast and actual data of wind and solar power output and runoff, calculate forecast error, and count the mean and covariance matrix of forecast error in each period; (1.2) Error modeling and sampling: Generate random error samples using multivariate normal distribution based on the error mean and covariance matrix; (1.3) Scenario generation: The prediction error is superimposed on the random error sample to generate a water-wind-light joint scenario; Step (2) establishes a short-term joint scheduling model for water, wind and solar power as the first stage model. The model is as follows: Objective function: Where: ns is the number of scenes; s is the scene number; t is the time period number; short t,s is the power shortage output of scenario s in period t, MW; curtail t,s is the curtailed power output of scenario s in period t, MW; The constraints include hydraulic constraints and power constraints. The hydraulic constraints include water balance constraints, water level control constraints at the beginning and end of the dispatch period, upper and lower limits of water level, water level and reservoir capacity relationship, upper and lower limits of power generation flow, upper and lower limits of outflow flow, and upper and lower limits of power station output. The power constraints include power balance constraints and transmission line constraints. Step (3) uses the water-wind-light joint scene generated in step (1) as input conditions, solves the stage one model in (2), and obtains the input parameters of the stage two model. The specific steps are as follows: (3.1) Considering the current and future scenarios of two new energy installed capacity, the water-wind-solar joint scenario is generated by the method in step (1), and the generated scenario is used as the input condition of the stage 1 model. The stage 1 model is solved using the commercial solver gurobi to obtain the output process of each power source and the load tracking situation; (3.2) Based on the load tracking situation under the future wind and solar power installed capacity scenario calculated in step (3.1), evaluate the power abandonment and power shortage in this scenario; combine the power abandonment and power shortage, take the larger of the two, and obtain the energy storage scale required to achieve complete power transfer, and use this energy storage scale as the upper limit of the energy storage capacity in stage 2; Step (4) establishes a hydropower, wind, solar and energy storage joint scheduling model as the stage 2 model. The objective function and hydropower constraints and power constraints are the same as those of the stage 1 model. The energy storage constraints are as follows: Where: r 0 Represents the energy storage charging state variable, which is a 0-1 variable; r 1 Indicates the energy storage discharge state variable, which is a 0-1 variable; represents the output of energy storage in scenario s at time period t, MW; represents the power generation of energy storage in scenario s at time period t, MW; represents the charging power of the energy storage in scenario s at time period t; Indicates the maximum discharge power of energy storage; Indicates the maximum charging power of the energy storage; in this calculation, the reduction of the upper limit of the energy storage discharge power as the battery power decreases is not considered; The energy storage utilization index is used to evaluate the utilization of energy storage and analyze the rationality of configuring energy storage of different scales. The index calculation formula is shown in formula (6); Where: UR is the energy storage utilization rate; Represents the average value of energy storage charging / discharging power, MW; P max Indicates the maximum value of the configured energy storage charging and discharging power; Step (5) is to solve the model of stage 2 according to the upper limit of energy storage scale solved in stage 1 in step (3), and to calculate the energy storage utilization index by giving energy storage of different scales as input conditions. The specific steps are as follows: (5.1) Two energy storage scale scenarios are considered: one is to configure energy storage based on the larger of the upper limits of power shortage or power abandonment, and the other is to configure energy storage based on the average power abandonment output and the average power shortage output; (5.2) Use the commercial solver gurobi to solve the stage 2 model and obtain the output process of each power source, the energy storage charging and discharging process, and the load tracking situation; (5.3) According to the energy storage charging and discharging process, the energy storage utilization rate UR is calculated according to formula (6). By comparing the effects of different scales of energy storage on reducing power shortages and power abandonment in the power grid, the construction plan of supporting energy storage on the grid side in the future is determined.

2. A two-stage optimization method for short-term coordinated scheduling of water, wind, solar and energy storage according to claim 1, characterized in that: The constraints in step (2) are as follows: Hydropower constraints: Where: V m,t+1 and V m,t are the water storage capacity (storage capacity) of hydropower station m at the beginning of time period t+1 and t, respectively. 3 ;Q m,t and S m,t are the inflow and outflow of reservoir m in period t, respectively. 3 / s; Δt is the calculation period step; Z m,0 represents the starting water level of the dispatching period of hydropower station m, m; Z m,T represents the end water level of the dispatch period of hydropower station m, m; Z m,t and They represent the water level Z of hydropower station m at time t. m,t The lower and upper limits of are taken as dead water level to normal high water level in the non-flood season and dead water level to flood limit water level in the flood season; f represents the water level and storage capacity relationship of reservoir m; q m,t and They represent the power generation flow q of hydropower station m in period t respectively. m,t The lower and upper limits of m 3 / s;S m,t and They represent the outflow S of hydropower station m in period t. m,t The lower and upper limits of m 3 / s; and They represent the output of hydropower station m in period t. The lower and upper limits of MW; Power Constraints: Where: Load is the total load of various power sources in the dispatching area, including external load, MW; The output of all hydropower stations at time t in scenario s, MW; The output of all wind power plants at time t in scenario s, MW; Output of all PV power stations at time t in scenario s, MW; P l,t,s is the channel output of line l in scenario s at time period t, MW; is the output of hydropower station m at time t in scenario s; M is the number of hydropower stations in the calculation area; M l To calculate the number of hydropower stations in the regional channel l; is the sum of the output of other power sources except hydropower in line l in scenario s and time period t; P l,max is the maximum transmission capacity of channel l; different power stations in cascade hydropower may transmit electricity through different lines.

Images