Method and device for optimizing peak regulation capacity demand of provincial power system

Abstract

The invention provides a peak regulation capacity demand optimization method and device for a provincial power system. The method comprises the steps of obtaining a day-ahead operation scene of the provincial power system; obtaining an internal peak regulation resource model of the provincial power system, establishing an energy storage scheduling model according to the internal peak regulation resource model of the provincial power system, and solving the energy storage scheduling model according to the day-ahead operation scene of the provincial power system to obtain a resource scheduling scheme; and optimizing the peak regulation capacity of the provincial power system by considering multiple uncertainties, and screening the optimal peak regulation capacity of the provincial power system in the resource scheduling scheme according to the power supply cost. According to the technical scheme provided by the invention, the peak regulation demand capacity of the provincial system can be optimized by considering the strong uncertainty of the renewable energy output and the load at the same time, and the problem that the peak regulation pressure of the power grid is increased due to the strong uncertainty of the renewable energy output and a relatively large prediction error is solved.

Description

Technical Field

[0001] This document relates to the technical field of peak-shaving capacity demand of power systems, and in particular to a method and system for optimizing peak-shaving capacity demand of provincial power systems. Background Art

[0002] my country's installed capacity of new energy will continue to expand, but the randomness of renewable energy has increased the peak-shaving pressure on the power grid. Traditional thermal power units are still the main means of peak-shaving, and the increase in the proportion of new energy has made it even more burdensome. Coordinating various peak-shaving resources to solve the problem of insufficient peak-shaving capacity has become a research hotspot. If the system can take appropriate peak-shaving measures or install equipment that can cooperate with peak-shaving, it can effectively compensate for the intermittent, variable and uncertain output of wind turbines and photovoltaics. From the load side, demand response can guide users' electricity consumption behavior through real-time electricity prices in the power market, give full play to the peak-shaving capacity of the load side, and reduce the peak-shaving pressure of the system. However, the demand response resources in the power system are often limited, and it is difficult to solve the problem of insufficient peak-shaving capacity of the new power system caused by large-scale new energy grid connection. Inter-provincial interconnection lines have the characteristics of large capacity, controllability and flexibility, and have a wide range of applications. Optimizing the power purchase / sale plan of provincial interconnection lines for the peak-shaving power demand of provincial power systems can make full use of the resource endowments of various regions, while meeting the stable operation of the power system, and reducing the power supply cost of the system.

[0003] However, there are certain errors in the forecast of renewable energy output, which makes it impossible to optimize the peak-shaving power of interconnection lines based entirely on the forecast output of renewable energy. At present, the main methods for dealing with the uncertainty of renewable energy power forecast are: (1) increasing spinning reserve. This method is simple and reliable, but because the current forecast error of wind power is still large, it is difficult to determine the size of the reserved spinning reserve capacity; (2) robust optimization method. The idea of ​​robust optimization is to transform the uncertainty of wind power into the form of uncertain solution set. Considering the worst scenario, it often makes the dispatch result conservative and less economical; (3) random modeling method based on chance-constrained programming. This method is difficult to derive an analytical chance-constrained model for large-scale power systems, and it is also difficult to objectively determine what confidence level to set; (4) random optimization method based on scenario set. Compared with the previous methods, this method can avoid overly conservative dispatch results and improve the rationality of the optimization scheme when taking into account the probability distribution of renewable energy output. It is suitable for the peak-shaving capacity demand optimization method of provincial power systems with high penetration of renewable energy.

[0004] In summary, in order to solve the problem of strong uncertainty in renewable energy output and large prediction errors leading to increased peak-shaving pressure on the power grid, a provincial power system peak-shaving capacity demand optimization method that considers multiple uncertainties is urgently needed. Summary of the invention

[0005] The present invention provides a method and system for optimizing the peak-shaving capacity demand of a provincial power system, which are used to solve the problem of increased peak-shaving pressure on a power grid caused by strong uncertainty in the output of renewable energy and large prediction errors.

[0006] According to an embodiment of the present invention, a method for optimizing peak load capacity demand of a provincial power system is provided, comprising:

[0007] S1. Fitting the probability distribution function of renewable energy and load output forecast error based on moment estimation method, and using Latin hypercube sampling to generate the day-ahead operation scenario of the provincial power system;

[0008] S2. Obtaining a provincial power system internal peak-shaving resource model, wherein the provincial power system internal peak-shaving resource model includes: establishing a demand-side response resource scheduling model, establishing a thermal power unit scheduling model, establishing an energy storage scheduling model, and establishing a hydropower unit scheduling model;

[0009] S3. Establish an energy storage dispatch model based on the internal peak load resource model of the provincial power system, and solve the energy storage dispatch model to obtain a resource dispatch plan based on the day-ahead operation scenario of the provincial power system;

[0010] S4. Optimize the peak-shaving capacity of the provincial power system by considering multiple uncertainties, and select the best peak-shaving capacity of the provincial power system in the resource scheduling scheme according to the power supply cost.

[0011] According to an embodiment of the present invention, a device for optimizing peak load capacity demand of a provincial power system is provided, comprising:

[0012] The scenario acquisition module fits the probability distribution function of renewable energy and load output forecast errors based on the moment estimation method, and uses Latin hypercube sampling to generate the day-ahead operation scenario of the provincial power system;

[0013] A peak-shaving resource model acquisition module is used to acquire a peak-shaving resource model within a provincial power system, wherein the peak-shaving resource model within a provincial power system includes: establishing a demand-side response resource scheduling model, establishing a thermal power unit scheduling model, establishing an energy storage scheduling model, and establishing a hydropower unit scheduling model;

[0014] A simulation model module, which establishes an energy storage dispatching model according to the internal peak-shaving resource model of the provincial power system, and solves the energy storage dispatching model according to the day-ahead operation scenario of the provincial power system to obtain a resource dispatching plan;

[0015] The optimal peak-shaving capacity module is obtained, the peak-shaving capacity of the provincial power system is optimized considering multiple uncertainties, and the optimal peak-shaving capacity of the provincial power system is selected in the resource scheduling scheme according to the power supply cost.

[0016] The embodiment of the present invention takes the optimal scheduling of the peak-shaving demand capacity of the provincial power system as the research object. By adopting the optimization method of the peak-shaving capacity demand of the provincial power system of the embodiment of the present invention, it is only necessary to input the easily accessible load data of the local power system and the output data of renewable energy to carry out the optimal scheduling of the peak-shaving demand capacity. The embodiment of the present invention simultaneously considers the prediction errors of renewable energy output and load, and comprehensively utilizes the demand-side response resources and the inter-provincial interconnection line power grid resources to establish an optimal scheduling model for the peak-shaving demand capacity. The linear programming method can be used to quickly solve the power supply cost. Within the feasible solution interval of the peak-shaving demand capacity, a one-dimensional traversal method is used to search, and the optimal peak-shaving capacity is selected according to the expected power supply cost, which has good engineering application value. It solves the problem of increased peak-shaving pressure on the power grid caused by the strong uncertainty of renewable energy output and large prediction errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0018] Figure 1 It is a flow chart of a method for optimizing the peak load capacity demand of a provincial power system according to an embodiment of the present invention;

[0019] Figure 2 A schematic diagram of a device for optimizing peak load capacity demand of a provincial power system according to an embodiment of the present invention;

[0020] Figure 3 It is a detailed flow chart of the method for optimizing the peak load capacity demand of the provincial power system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0022] Method Embodiment

[0023] According to an embodiment of the present invention, a method for optimizing peak load capacity demand of a provincial power system is provided. Figure 1 is a flow chart of a method for optimizing the peak load capacity demand of a provincial power system according to an embodiment of the present invention, Figure 3 is a detailed flow chart of the method for optimizing the peak load capacity demand of the provincial power system according to an embodiment of the present invention. Figure 1 and Figure 3 As shown, the provincial power system peak load capacity demand optimization method of the embodiment of the present invention specifically includes:

[0024] S1. Fitting the probability distribution function of renewable energy and load output forecast error based on moment estimation method, and using Latin hypercube sampling to generate the day-ahead operation scenario of the provincial power system;

[0025] The forecast error of renewable energy output at each time section is zero and the standard deviation is σ t The normal distribution of renewable energy output forecast error is σ t Calculated by the following formula:

[0026]

[0027] in, P cap They represent the output of renewable energy and the installed capacity of renewable energy at time t respectively.

[0028] However, there is no uniformly recognized distribution function for load forecast error distribution in the industry. Therefore, it is necessary to fit the forecast error according to the forecast error distribution law, and combine the verified forecast error statistical distribution law with the deterministic load forecast result to obtain the load forecast result. In fact, according to the statistical results of the data, it is found that the standard deviation of the forecast error for each time section is σ t It is not a fixed ratio with the output of renewable energy. The longer the prediction time scale, the lower the prediction accuracy tends to be.

[0029] Therefore, the embodiment of the present invention uses the historical record data of the past year and the moment estimation method to obtain the standard deviation of each time section. The historical measurement data of renewable energy at the tth moment of each day is recorded as The historical forecast data of renewable energy at time t every day is recorded as Then we can get the deviation rate of the forecast error at the tth moment of each day, recorded as X t , X t =X 1,t X 2,t ...X 365,t ], can be calculated by the following formula:

[0030]

[0031] Among them, . / represents the ratio of corresponding elements of two vectors.

[0032] The sample mean at time t and the sample second-order central moment M t As the population mean μ t and variance The moment estimate of is calculated as follows:

[0033]

[0034] Then the normal distribution parameters of the deviation rate of the prediction error of each time section can be obtained, that is, N(μ t , ).

[0035] The use of Latin hypercube sampling to generate the day-ahead operation scenario of the provincial power system specifically includes:

[0036] Latin hypercube sampling is a stratified sampling method that consists of two steps: generating the original scene using Latin hypercube sampling, the specific steps are as follows:

[0037] The sampling process includes: assuming x1, x2, ..., x s There are s input random variables, where the kth random variable x k The cumulative probability distribution function of (k=1,2,…,s) is:

[0038] μ k =F k (x k );

[0039] Assume the sampling scale is N, and set F k The value space of x (for example, 0-1) is evenly divided into N intervals. A number is randomly selected from each interval. Then the variable x k The sample value of can be calculated using the inverse function method as follows:

[0040] x ki =F k -1 ((i-η) / N);

[0041] Where: x ki is a random variable x k The sample in the i-th interval; η is a random number between 0 and 1; i = 1, 2, ..., N. k The sample values ​​of X are arranged in a row, and we can get k ; Sampling all input random variables can obtain an s×N dimensional sample matrix X.

[0042] Arrangement process:

[0043] After obtaining the sample matrix X, it is necessary to sort the positions of the elements in each row of X. The most basic arrangement method is random arrangement. For independent random variables, the correlation coefficient matrix of its sample matrix should be a unit matrix. If RP is used for sorting, the correlation coefficient matrix of the sample matrix may become a non-unit matrix. For this reason, a more reasonable arrangement method is needed to eliminate the unexpected correlation. Cholesky decomposition has high computational efficiency and excellent correlation control effect. The process of the arrangement method based on Cholesky decomposition is as follows:

[0044] Assume that L is an s×N dimensional sequential matrix, each row of which is a certain arrangement of 1, 2, …, N, which corresponds to the arrangement of each row element of the sample matrix X. A is an s×s dimensional linear correlation matrix of L, which is a pairwise positive definite matrix, so A can be decomposed by Cholesky:

[0045] A=DD T ;

[0046] Where: D is a lower triangular matrix.

[0047] The s×N dimensional matrix G can be calculated as follows:

[0048] G=D -1 L;

[0049] Unlike matrix L, the elements of matrix G are not necessarily positive integers. Generate matrix G' based on G, so that each row of G' is a permutation of 1, 2, ..., N, and its permutation order corresponds to the permutation order of each row of elements in G. According to the permutation of each row of elements in G', update the permutation of each row of elements in sample matrix X to obtain the final sample matrix X'.

[0050] Generate renewable energy and load output scenario sets based on the adoption process and arrangement process, including:

[0051] According to the above sampling process and arrangement process, the prediction error rate of J scenarios is generated, where the prediction error rate of the jth scenario is recorded as X j , The predicted output vector of wind turbines, photovoltaics and the next dispatching period is denoted as P PV , P Wind , P L , which are: The output of the photovoltaic meter and the prediction error in the jth scenario can be calculated using the following formula:

[0052] P PV =P PV +P PV *X PV,j

[0053] Similarly, the output of the wind turbine and load in the jth scenario taking into account the prediction error can be calculated.

[0054] S2. Obtaining a provincial power system internal peak-shaving resource model, wherein the provincial power system internal peak-shaving resource model includes: establishing a demand-side response resource scheduling model, establishing a thermal power unit scheduling model, establishing an energy storage scheduling model, and establishing a hydropower unit scheduling model;

[0055] The establishment of a demand-side response resource scheduling model includes:

[0056] Calculate the objective function: obtain the demand side response call cost according to the unit compensation electricity price of the user who performs the demand side response, the original load of the user who does not participate in the demand side response, and the new load power of the user who participates in the demand side response;

[0057] More and more users have the ability and willingness to participate in demand-side response. Transferable load demand-side response is introduced. The demand-side response call cost is denoted as C1, and the calculation method is shown in the following formula.

[0058]

[0059] Among them, C t,DR Represents the unit compensation electricity price for users who perform demand-side response at time t; represents the original load of users who do not participate in demand-side response at time t; Represents the new load power of users participating in demand-side response at time t.

[0060] Considering the constraints: Obtain the demand-side response constraints of the provincial power system according to the load transfer depth when all participating loads are transferable loads and there are restrictions on the transfer depth of the loads, including:

[0061] Assuming that all the loads participating in the demand-side response of the provincial power system are currently transferable loads and there are restrictions on the transfer depth of the loads, the following demand-side response constraints are established.

[0062]

[0063] Wherein, k1 and k2 represent the load transfer depths respectively.

[0064] The establishment of the thermal power unit dispatching model comprises:

[0065] Calculate the objective function: The operating cost of the thermal power unit is obtained according to the operating cost coefficient of the thermal power unit, the active output and start-stop cost of the thermal power unit, the start-stop status of the thermal power unit, and the start-stop cost of the unit; the operating cost of the thermal power unit is obtained by the following formula:

[0066]

[0067] Among them, a i , b i and c i is the operating cost coefficient of the i-th thermal power unit; and They represent the active power output and start-up and shutdown costs of the i-th thermal power unit at time t; S i,t It indicates the start and stop status of the i-th thermal power unit at time t, where start is 1 and stop is 0. Represents the start-up and shutdown cost of the unit.

[0068] Considered constraints: Establish startup and shutdown time constraints for thermal power units, establish minimum output, maximum output, maximum ramp speed, minimum startup time, and minimum shutdown time constraints for thermal power units, and obtain them through the following formulas:

[0069]

[0070] and They represent the startup and shutdown duration of the i-th thermal power unit at time t respectively; and They represent the minimum output, maximum output, maximum ramp speed, minimum startup time and minimum shutdown time of the i-th unit respectively;

[0071] The establishing of the energy storage scheduling model comprises:

[0072] Calculate the objective function: Obtain the energy storage operation cost based on the cost of each 1kWh of energy storage equipment discharged and the discharge power of the energy storage equipment; the energy storage operation cost can be calculated using the following formula:

[0073]

[0074] Among them, s ES P represents the cost of energy storage equipment per 1kWh of discharge; t Dis represents the discharge power of the energy storage device at time t;

[0075] The constraints taken into account include:

[0076] The relationship constraint between the energy storage capacity and the charging and discharging power is obtained through the energy storage equipment's initial time and the end time of the scheduling cycle, the energy storage equipment's capacity and the charging power; The relationship constraint between the energy storage capacity and the charging and discharging power is:

[0077]

[0078] Among them, E1 and E LastThey represent the power of the energy storage device at the initial moment and at the end of the dispatch cycle respectively; E t and P t Cha Indicates the power of the energy storage device and the charging power at time t.

[0079] The upper and lower limits of energy storage capacity are obtained by obtaining the maximum power of the energy storage device; the energy storage capacity has upper and lower limits:

[0080]

[0081] Among them, E max Indicates the maximum power of the energy storage device. m1 and m2 are constants and are related to the type of energy storage device.

[0082] There is an upper limit constraint on obtaining the charging and discharging power of the energy storage device, and the constraint that the energy storage device is not allowed to charge and discharge at the same time.

[0083] There is an upper limit constraint on the charging and discharging power of energy storage equipment:

[0084]

[0085] Among them, λ is a constant, which is related to the energy storage device used.

[0086] Energy storage devices are not allowed to charge and discharge simultaneously:

[0087] P t Dis *P t Cha =0;

[0088] The establishment of the hydropower unit dispatching model comprises:

[0089] The operating cost of the hydropower unit is negligible, and the following constraints need to be taken into account:

[0090]

[0091] Among them, P t W , P t W,min and E W They respectively represent the output of the hydropower unit, the minimum output and the amount of electricity that can be provided within a scheduling cycle.

[0092] S3: Establishing an energy storage dispatching model based on the internal peak load resource model of the provincial power system, and solving the energy storage dispatching model to obtain a resource dispatching plan based on the day-ahead operation scenario of the provincial power system. S3 specifically includes:

[0093] The peak-shaving cost is obtained through the real-time electricity price of the peak-shaving capacity and peak-shaving electricity purchased by the power grid from the upper-level power grid and the peak-shaving capacity and peak-shaving electricity purchased by the power grid from the upper-level power grid;

[0094] The regional power grid provides peak load capacity and power for the provincial power system. The peak load cost C4 can be calculated as follows:

[0095]

[0096] in, and They respectively represent the real-time electricity prices of the peak-shaving capacity and peak-shaving electricity purchased by the power grid from the upper-level power grid; and It indicates the peak-shaving capacity and peak-shaving electricity purchased by the power grid from the superior power grid.

[0097] The objective function of the energy storage dispatch model is obtained based on the demand-side response call cost, the operating cost of thermal power units, the energy storage operating cost, and the peak-shaving cost. The objective function of the provincial power system day-ahead dispatch model can be expressed as follows:

[0098] C=C1+C2+C3+C4;

[0099] Obtaining constraints of the energy storage scheduling model, wherein the constraints of the energy storage scheduling model include: power balance constraints, peak-shaving power upper limit constraints, obtaining wind turbines, photovoltaic output upper limit constraints, spinning reserve constraints, and peak-shaving power upper and lower limit constraints;

[0100] Power balance constraints:

[0101]

[0102] Among them, P t L represents the load of the power grid at time t; P t dis and P t cha Represents the charging and discharging power of the grid energy storage device, P t PV and P t Wind Indicates the output level of photovoltaic and wind turbines in the corresponding scenario.

[0103] The upper limit constraints of peak load power are as follows:

[0104]

[0105] Wind turbine and photovoltaic output upper limit constraints:

[0106]

[0107] The spinning reserve constraints are as follows:

[0108]

[0109] Peak load power upper and lower limits constraints:

[0110]

[0111] Solving the objective function based on the constraint conditions of the energy storage scheduling model to obtain a resource scheduling solution specifically includes:

[0112] C2 is linearized using the linearization method, and the cost of calling demand-side response resources can be used as follows:

[0113] Introduce a new variable u i ' ,t 、v i ' ,t and w i ' ,t Let C1 be linearized, and let:

[0114]

[0115] The user compensation cost of the above demand-side response resource is converted into C3':

[0116]

[0117] S4. Optimizing the peak-shaving capacity of the provincial power system by considering multiple uncertainties, and selecting the best peak-shaving capacity of the provincial power system in the resource scheduling scheme according to the power supply cost. S4 specifically includes:

[0118] Input the feasible solution interval of provincial power system peak load capacity

[0119] Within the feasible solution interval of the peak load capacity of the peak power system, the one-dimensional traversal method is used to search and select the optimal peak load capacity according to the expected power supply cost, including:

[0120] Will Substitute the jth scenario among the J scenarios generated in step 1, and use step 3 to establish and solve the objective function of the provincial power system under the peak-shaving capacity. A total of J values ​​can be obtained, and their expected values ​​are used as the power supply cost of the provincial power system corresponding to the peak-shaving capacity;

[0121] make calculate The corresponding provincial power system power supply cost is

[0122]

[0123] The power supply cost corresponding to the peak-shaving capacity is screened, and the one with the lowest power supply cost is the optimal peak-shaving capacity of the provincial power system.

[0124] The embodiment of the present invention optimizes the peak-shaving demand capacity of the provincial system by considering the strong uncertainty of renewable energy output and load. A method for fitting the probability distribution function of renewable energy and load output forecast errors based on the moment estimation method is proposed, so that the optimization scheduling model can effectively take into account the forecast errors of renewable energy output and load; a provincial power system peak-shaving capacity demand optimization model is established, and a corresponding solution method is given.

[0125] By adopting the embodiments of the present invention, the following beneficial effects are achieved:

[0126] The embodiment of the present invention takes the optimal scheduling of the peak-shaving demand capacity of the provincial power system as the research object. By adopting the optimization method of the peak-shaving capacity demand of the provincial power system of the embodiment of the present invention, it is only necessary to input the easily accessible load data of the local power system and the output data of renewable energy to carry out the optimal scheduling of the peak-shaving demand capacity. The embodiment of the present invention simultaneously considers the prediction errors of renewable energy output and load, and comprehensively utilizes the demand-side response resources and the inter-provincial interconnection line power grid resources to establish an optimal scheduling model for the peak-shaving demand capacity. The linear programming method can be used to quickly solve the power supply cost. Within the feasible solution interval of the peak-shaving demand capacity, a one-dimensional traversal method is used to search, and the optimal peak-shaving capacity is selected according to the expected power supply cost, which has good engineering application value. It solves the problem of increased peak-shaving pressure on the power grid caused by the strong uncertainty of renewable energy output and large prediction errors.

[0127] Device Embodiment

[0128] According to an embodiment of the present invention, a device for optimizing peak load capacity demand of a provincial power system is provided. Figure 2 Schematic diagram of a device for optimizing the peak load capacity demand of a provincial power system according to an embodiment of the present invention. Figure 2 As shown, the provincial power system peak load capacity demand optimization device of the embodiment of the present invention specifically includes:

[0129] The scenario acquisition module 20 fits the probability distribution function of renewable energy and load output forecast errors based on the moment estimation method, and generates the day-ahead operation scenario of the provincial power system using Latin hypercube sampling;

[0130] The peak-shaving resource model acquisition module 22 acquires the internal peak-shaving resource model of the provincial power system, wherein the internal peak-shaving resource model of the provincial power system includes: establishing a demand-side response resource scheduling model, establishing a thermal power unit scheduling model, establishing an energy storage scheduling model, and establishing a hydropower unit scheduling model;

[0131] The simulation model module 24 establishes an energy storage dispatching model according to the internal peak load resource model of the provincial power system, and solves the energy storage dispatching model according to the day-ahead operation scenario of the provincial power system to obtain a resource dispatching plan;

[0132] The module 26 for obtaining the optimal peak-shaving capacity optimizes the peak-shaving capacity of the provincial power system by considering multiple uncertainties, and selects the optimal peak-shaving capacity of the provincial power system in the resource scheduling scheme according to the power supply cost.

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A provincial power system peak load capacity demand optimization method, characterized in that include: S1. Fitting the probability distribution function of renewable energy and load output forecast error based on moment estimation method, and using Latin hypercube sampling to generate the day-ahead operation scenario of the provincial power system; S2. Obtaining a provincial power system internal peak-shaving resource model, wherein the provincial power system internal peak-shaving resource model includes: establishing a demand-side response resource scheduling model, establishing a thermal power unit scheduling model, establishing an energy storage scheduling model, and establishing a hydropower unit scheduling model; S3. Establish an energy storage dispatch model based on the internal peak load resource model of the provincial power system, and solve the energy storage dispatch model to obtain a resource dispatch plan based on the day-ahead operation scenario of the provincial power system; S4. Optimize the peak-shaving capacity of the provincial power system by considering multiple uncertainties, and select the best peak-shaving capacity of the provincial power system in the resource scheduling scheme according to the power supply cost.

2. The method according to claim 1, characterized in that The method of fitting the probability distribution function of renewable energy and load output prediction error based on moment estimation method specifically includes: Obtain historical data within one year and use moment estimation method to obtain the standard deviation of each time section; the historical measurement data of renewable energy at the tth moment of each day is recorded as P t true , The historical forecast data of renewable energy at time t every day is recorded as P t fore , According to P t true and P t fore Get the deviation rate of the forecast error at time t every day, recorded as X t , X t =[X 1,t X 2,t ...X 365,t ], X t Obtained by formula 1: X t =(P t true -P t fore ). / P t true *100% Formula 1; Among them, . / represents the ratio of the corresponding elements of two vectors; The sample mean at time t and the sample second-order central moment M t As the population mean μ t and variance The moment estimate of is calculated as follows: Get the normal distribution parameters of the deviation rate of the prediction error of each time section, and then get the error probability distribution function 3. The method according to claim 1, characterized in that The use of Latin hypercube sampling to generate the day-ahead operation scenario of the provincial power system specifically includes: Obtain an s×N-dimensional sample matrix X, perform Cholesky decomposition on the sample matrix X, and generate the prediction error rate of J scenarios, where the prediction error rate of the jth scenario is recorded as X j , The predicted output vector of wind turbines, photovoltaics and loads in the next dispatch period is denoted as P PV , P Wind , P L , which are: The output of the photovoltaic meter and the prediction error in the jth scenario can be calculated using the following formula: P PV =P PV +P PV *X PV,j Formula 3; Based on the same method, the output of the wind turbine and load in the jth scenario taking into account the prediction error is obtained.

4. The method according to claim 1, characterized in that: The establishment of a demand-side response resource scheduling model includes: Obtaining the demand side response calling cost according to the unit compensation electricity price for the user who performs the demand side response, the original load of the user who does not participate in the demand side response, and the new load power of the user who participates in the demand side response; According to the load transfer depth, the demand side response constraint conditions of the provincial power system are obtained when all the participating loads are transferable loads and there are restrictions on the transfer depth of the loads.

5. The method according to claim 1, characterized in that: The establishment of the thermal power unit dispatching model comprises: The operation cost of the thermal power unit is obtained according to the operation cost coefficient of the thermal power unit, the active output and start-stop cost of the thermal power unit, the start-stop status of the thermal power unit and the start-stop cost of the unit; Establish startup and shutdown time constraints for thermal power units, and establish minimum output, maximum output, maximum climbing speed, minimum startup time and minimum shutdown time constraints for thermal power units.

6. The method according to claim 1, characterized in that The establishing of the energy storage scheduling model comprises: The energy storage operation cost is obtained based on the cost of each 1kWh of energy storage equipment discharged and the discharge power of the energy storage equipment; Obtain the relationship constraint between the energy storage power and the charging and discharging power through the power of the energy storage device at the initial moment and the end moment of the scheduling cycle, the power of the energy storage device and the charging power; Obtain the upper and lower limits of energy storage capacity through the maximum capacity of energy storage equipment; There is an upper limit constraint on obtaining the charging and discharging power of the energy storage device, and the constraint that the energy storage device is not allowed to charge and discharge at the same time.

7. The method according to claim 1, characterized in that The establishment of the hydropower unit dispatching model comprises: The scheduling constraints of the hydropower units are obtained through the output, minimum output and the amount of electricity that can be provided within a scheduling cycle.

8. The method according to claim 1, characterized in that: The S3 specifically includes: The peak-shaving cost is obtained through the real-time electricity price of the peak-shaving capacity and peak-shaving electricity purchased by the power grid from the upper-level power grid and the peak-shaving capacity and peak-shaving electricity purchased by the power grid from the upper-level power grid; Obtain the objective function of the energy storage dispatch model based on the demand-side response call cost, the operating cost of thermal power units, the energy storage operating cost, and the peak load cost; Obtaining constraints of the energy storage scheduling model, wherein the constraints of the energy storage scheduling model include: power balance constraints, peak-shaving power upper limit constraints, obtaining wind turbines, photovoltaic output upper limit constraints, spinning reserve constraints, and peak-shaving power upper and lower limit constraints; The objective function is solved based on the constraints of the energy storage scheduling model to obtain a resource scheduling solution.

9. The method according to claim 1, characterized in that: The S4 specifically includes: Obtain feasible solution interval for peak load regulation capacity of provincial power system Within the feasible solution interval of the peak-shaving capacity of the peak power system, a one-dimensional traversal method is used to search and select the optimal peak-shaving capacity according to the expected power supply cost.

10. A device for optimizing peak load capacity demand of a provincial power system, characterized in that: include: The scenario acquisition module fits the probability distribution function of renewable energy and load output forecast errors based on the moment estimation method, and uses Latin hypercube sampling to generate the day-ahead operation scenario of the provincial power system; A peak-shaving resource model acquisition module is used to acquire a peak-shaving resource model within a provincial power system, wherein the peak-shaving resource model within a provincial power system includes: establishing a demand-side response resource scheduling model, establishing a thermal power unit scheduling model, establishing an energy storage scheduling model, and establishing a hydropower unit scheduling model; A simulation model module, which establishes an energy storage dispatching model according to the internal peak-shaving resource model of the provincial power system, and solves the energy storage dispatching model according to the day-ahead operation scenario of the provincial power system to obtain a resource dispatching plan; The optimal peak-shaving capacity module is obtained, the peak-shaving capacity of the provincial power system is optimized considering multiple uncertainties, and the optimal peak-shaving capacity of the provincial power system is selected in the resource scheduling scheme according to the power supply cost.

Images