Wind storage system robust-random decision control method considering internal and external uncertain factors

Abstract

The invention discloses a wind storage system robust-random decision control method considering internal and external uncertain factors. The method comprises the following steps: firstly, determining an energy consumption objective function of the wind storage system according to uncertain factors and internal composition factors of the wind storage system, and establishing an energy consumption model of the wind storage system; then determining a clearing target function according to factors except the wind storage system and establishing a clearing model; establishing a wind storage system double-layer model according to the obtained energy consumption model and the obtained clearing model; and finally, optimizing and solving the double-layer model of the wind storage system by adopting an embedded Carlo demand-Kuhn-tuck condition method and an improved constraint and column generation algorithm in sequence to obtain the lowest total energy consumption of the wind storage system. The method is comprehensive in consideration and sufficient in coverage factor, so that the operation parameter evaluation is more practical, and the obtained minimum energy consumption can bring the maximum power storage value.

Description

Technical Field

[0001] The invention relates to a method for minimizing wind energy consumption of a storage system, and in particular to a robust-random decision control method for a wind energy storage system taking internal and external uncertain factors into consideration. Background Art

[0002] With the continuous deepening of power system reform, energy conservation and carbon reduction requirements have become increasingly prominent, and the proportion of renewable energy in total power generation has increased year by year. Wind power generation systems usually need to be equipped with energy storage facilities to form a wind storage system. This system can not only effectively absorb wind power generation, but also utilize energy storage characteristics. However, the wind storage system faces many types of uncertain factors and is affected by the fluctuations of wind power output itself. The interweaving of multiple uncertainties will inevitably affect the energy consumption of the wind storage system. Therefore, how to determine the scientific and reasonable energy consumption cost of the wind storage system has important research value and practical significance. Summary of the invention

[0003] In order to solve the problems existing in the background technology, the present invention provides a robust-stochastic decision control method for a wind storage system taking into account internal and external uncertainties. By using a random-robust method, various internal and external uncertainties faced by the wind storage system are modeled to fully consider the impact of these uncertainties. The energy storage and wind power of the wind storage system are modeled, and the day-ahead clearing is modeled, and the optimal objective function and safe operation constraints of the two-layer model are determined. The model is converted and solved using the embedded Carlo-Kuhn-Tucker condition method and the improved constraint and column generation algorithm. The optimization problem is expressed using Matlab, and the solver is used to solve the minimum total energy consumption of the wind storage system, which is finally input into the power grid through operating parameter evaluation to achieve decision control.

[0004] The technical solution adopted by the present invention is:

[0005] A robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties of the present invention comprises the following steps:

[0006] S1. Determine the energy consumption objective function of the wind storage system based on the uncertain factors and internal components of the wind storage system and establish an energy consumption model of the wind storage system.

[0007] S2. Determine the clearing objective function and establish a clearing model based on factors other than the wind storage system.

[0008] S3. Establish a double-layer model of the wind-storage system according to the energy consumption model obtained in step S1 and the clearing model obtained in step S2.

[0009] S4. The built-in Carlo-Kuhn-Tucker condition method and the improved constraint and column generation algorithm are used in turn to optimize and solve the two-layer model of the wind-storage system established in step S3 to obtain the minimum total energy consumption of the wind-storage system.

[0010] The step S1 is specifically as follows:

[0011] S11. Modeling the uncertain factors outside the wind storage system, where the uncertain factors outside the wind storage system include output fluctuations and real-time energy consumption costs outside the wind storage system.

[0012] S12. Modeling the uncertain factors within the wind storage system, wherein the uncertain factors within the wind storage system include output fluctuations within the wind storage system.

[0013] S13. Modeling the internal components of the wind storage system, wherein the internal components of the wind storage system include the energy storage system and the wind turbine generator set of the wind storage system.

[0014] S14. According to the modeling of steps S11 to S13, determine the energy consumption target function of the wind-storage system and establish an energy consumption model of the wind-storage system.

[0015] The energy consumption objective function of the wind energy storage system in step S1 is specifically set according to the following formula:

[0016]

[0017] Where s represents the real-time scenario; t represents the time point; q represents the scenario where the day-ahead uncertainty is generated; w represents the number of the wind turbine in the wind storage; ρ q and ρ s represents the probability of a single scenario generated by two scenarios; represents the unit energy consumption cost of wind turbine abandonment, represents the clearing energy consumption cost of the node where the wind storage system is located on the day before; and They represent the day-ahead charging power of the wind-storage system, the day-ahead discharging power of the wind-storage system, the real-time output of internal wind power, and the difference between the real-time and day-ahead power of the wind-storage system; Indicates the actual predicted value of the real-time energy consumption cost scenario; Indicates the actual output upper limit of internal wind power; It represents the real-time clearing output value of internal wind power; maxmin represents the maximum and minimum value problem, that is, solving the optimal state under the worst condition.

[0018] The energy consumption model of the wind energy storage system in step S1 satisfies the following constraints:

[0019]

[0020]

[0021] Among them, wo and na represent the code of the external wind turbine and the node number of the wind storage connection respectively; Indicates the output upper limit of external wind power; Represents the historical forecast average value of external wind power output; ε wo,t,q represents the historical forecast error; represents the standard deviation of historical forecast errors; Represents the real-time forecast average; Represents the real-time prediction error; represents the normal distribution of standard deviation, N() represents the normal distribution; w represents the number of wind turbines in the wind storage; and A 0-1 variable indicating the upper and lower limits of the output of the internal wind turbines have been reached. 0 means the internal fan output has not reached the upper limit; 1 means the internal fan output has reached the upper limit. 0 means that the internal fan output has not reached the lower limit; 1 means the internal fan output has reached the lower limit; and Respectively represent the historical forecast average value of internal wind power output, the historical output upper limit and the historical output lower limit; Γ 1 , Γ 2 They represent the conservative coefficients of the wind turbine number dimension and the time dimension respectively, e represents the number of energy storage; T represents the number of the last time period; and Respectively represent the real-time charging power of the internal energy storage and the real-time discharging power of the internal energy storage; and They represent the maximum value of charging and discharging of energy storage respectively; An indicator variable representing charging and discharging; represents the stored energy at time t; represents the stored energy at time t-1; and Respectively represent the upper and lower limits of the stored energy; η ec and η ed Respectively represent the charging efficiency and discharging efficiency of energy storage; E e,0 represents the initial value of stored energy; Indicates the final value of the stored energy in the real-time state; and They represent the output of internal wind power, the charging power of energy storage, and the discharging power of energy storage in the day-ahead state respectively; Indicates the maximum output value of internal wind power; The energy stored in the energy storage representing the state of the day before; Indicates the energy stored at the end of the energy storage in the day-ahead state; The energy stored in the previous time period represents the energy stored in the energy storage of the day-ahead state; With O acmax They represent the charging energy consumption cost and the upper limit of the charging energy consumption cost of the wind storage system respectively; With O admax They represent the discharge energy consumption cost and the upper limit of the discharge energy consumption cost of the wind storage system respectively; and They represent the charging and discharging capacity of the wind storage system respectively. It is the state variable of charging and discharging of the wind storage system.

[0022] The clearing objective function in step S2 is specifically set according to the following formula:

[0023]

[0024] Among them, g represents the number of ordinary units; G represents the number of ordinary units; They represent the energy consumption costs of traditional units, loads and independent wind power stations respectively; Indicates the power output of ordinary units; Indicates the load clearing amount; It indicates the power cleared by external independent wind power on the previous day.

[0025] The clearing model established in step S2 satisfies the following constraints:

[0026]

[0027]

[0028] Where, d is the load number; Indicates the output upper limit of ordinary units; and denote the dual variables of the upper and lower constraints of the traditional unit respectively; and They represent the upper and lower limits of load clearing respectively; and The dual variables representing the upper and lower constraints of the loading respectively; and The upper and lower limits of wind-storage system charging and discharging and the dual variables of upper and lower limit constraints are represented respectively; represents the output of independent wind power plants; represents the output upper limit of external wind power, and are the dual variables representing the upper and lower limit constraints of the wind turbine output respectively; represents the phase angle of node n in the system, represents the phase angle of node m in the system; Indicates the transmission upper limit of branch n to m; Bnm Represents the nodal admittance matrix of the system; The dual variable representing the transmission upper limit constraint of transmission line n to m; and are the dual variables representing the upper and lower bound constraints on the phase angle of node n; represents the phase angle of the reference node; The dual variable representing the reference node phase angle constraint; d∈Ψ n represents the node to which the load is connected; g∈Ψ n Indicates the node connected to the common unit; a∈Ψ n represents the node connected to the wind storage system; w∈Ψ n Represents the node to which the external independent wind power is connected; m∈Θ n Indicates that m is a node connected to n; It represents the clearing unit energy consumption value of the power system at the day-ahead node.

[0029] The step S4 is specifically as follows:

[0030] S41. Optimize the double-layer model of the wind-storage system established in step S3 into a single-layer model using the embedded Carlo-Kuhn-Tucker condition method.

[0031] S42. Use an improved constraint and column generation algorithm to solve the single-layer model obtained in step S41.

[0032] S43. Obtain the minimum total energy consumption of the wind storage system based on the solution results.

[0033] The objective function of the single-layer model obtained in step S41 is specifically set according to the following formula:

[0034]

[0035] in, Indicates the real-time energy consumption cost; represents the day-ahead energy cost.

[0036] The single-layer model obtained in step S41 satisfies the following constraints:

[0037]

[0038]

[0039] Where n(m∈Θ) represents the node number; represents the day-ahead cleared output power of external independent wind power; Indicates the real-time output power of internal wind power; Indicates the clearing of the connection nodes of the common units on the previous day; It represents the clearing unit energy consumption value of the day-ahead load connection node; It represents the clearing unit energy consumption value of the node connected to the wind storage system on the previous day; It represents the clearing unit energy consumption value of the external independent wind power connection node on the day before; The day-ahead node clearing unit energy consumption value of a node in a branch of the power system; It represents the unit energy consumption value of another node in a branch of the power system cleared the day before; Indicates the unit energy consumption value reported by ordinary units; Indicates the reported unit energy consumption value of the load; Indicates the reported unit energy consumption value of the wind storage system charging; Indicates the reported unit energy consumption value of the wind storage system discharge; Indicates the reported unit energy consumption value of the external independent fan; Indicates the clearing load value of the load in the day-ahead state.

[0040] The step S42 is specifically as follows:

[0041] 1) Divide the objective function of the single-layer model obtained in step S41 into the main problem and sub-problems And add constraints The η q Represents a minimum total energy consumption parameter set; Represents the real-time k-th energy consumption cost.

[0042] 2) Set the upper bound of the objective function value to LB = -∞ and the lower bound to UB = ∞, and initialize the number of iterations k, the convergence limit parameter v, and the robust uncertainty parameter Preset the convergence limit parameter v.

[0043] 3) Use the solver to solve the main problem Solve and obtain the optimal day-ahead energy cost of the main problem and the optimal minimum total energy consumption parameter set and use the best day-ahead energy cost and the optimal minimum total energy consumption parameter set Update the lower bound to

[0044] 4) The optimal minimum total energy consumption parameter set of the main problem obtained in step 3) Substitute into the subproblem The solver is used to solve the optimal real-time energy consumption cost and utilize the best real-time energy cost Update the upper bound to

[0045] 5) According to the updated upper and lower bounds, determine Whether it meets:

[0046] If satisfied, stop the iteration and get the solution;

[0047] If not satisfied, repeat steps 3) to 5) and update the number of iterations k.

[0048] The innovation of the present invention is that it fully considers the uncertainty factors inside and outside the wind storage system, the internal component factors and the factors outside the wind storage system, and adopts the embedded Carlo-Kuhn-Tucker condition method and the improved constraint and column generation algorithm to achieve comprehensive consideration and full coverage of factors, so that the operating parameter evaluation is more in line with reality, and the minimum energy consumption obtained can bring the maximum power storage value.

[0049] The beneficial effects of the present invention are:

[0050] 1. The present invention fully considers the uncertainty factors inside and outside the wind storage system, the internal component factors and the factors outside the wind storage system, which reflects that the present invention is comprehensive and covers all factors, making the operating parameter evaluation more in line with reality, and the minimum energy consumption obtained can bring the maximum power storage value.

[0051] 2. The present invention fully considers the mutual influence between the wind storage system and clearing, so that the energy consumption of the wind storage system is minimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0053] The present invention is described in more detail below in conjunction with the accompanying drawings and embodiments, but the present invention is not limited thereto. For those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention. The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in the field.

[0054] like Figure 1 As shown, the embodiment of the present invention includes the following steps:

[0055] S1. Determine the energy consumption objective function of the wind storage system based on the uncertain factors and internal components of the wind storage system and establish an energy consumption model of the wind storage system.

[0056] S11. The Monte Carlo random scenario generation method is used to model the uncertain factors outside the wind storage system. The uncertain factors outside the wind storage system include the output fluctuation and real-time energy consumption cost outside the wind storage system.

[0057] S12. A robust method is used to model the uncertainties within the wind-storage system. The uncertainties within the wind-storage system include output fluctuations within the wind-storage system.

[0058] S13. Modeling the internal components of the wind storage system, which include the energy storage system and wind turbines of the wind storage system.

[0059] S14. According to the modeling of steps S11 to S13, determine the energy consumption target function of the wind-storage system and establish an energy consumption model of the wind-storage system.

[0060] The energy consumption objective function of the wind energy storage system is set according to the following formula:

[0061]

[0062] Where s represents the real-time scenario, S represents the number of real-time scenarios and s∈S; t represents the time point; q represents the scenario generated by the day-ahead uncertainty, Q represents the number of scenarios generated by the day-ahead uncertainty and q∈Q; w represents the number of wind turbines in the wind storage; ρ q and ρ s represents the probability of a single scenario generated by two scenarios; represents the unit energy consumption cost of wind turbine abandonment, represents the clearing energy consumption cost of the node where the wind storage system is located on the day before; and They represent the day-ahead charging power of the wind-storage system, the day-ahead discharging power of the wind-storage system, the real-time output of internal wind power, and the difference between the real-time and day-ahead power of the wind-storage system; Indicates the actual predicted value of the real-time energy consumption cost scenario; Indicates the actual output upper limit of internal wind power; It represents the real-time clearing output value of internal wind power; maxmin represents the maximum and minimum value problem, that is, solving the optimal state under the worst condition.

[0063] The energy consumption model of the wind storage system meets the following constraints:

[0064]

[0065]

[0066]

[0067] Among them, wo and na represent the code of the external wind turbine and the node number of the wind storage connection respectively; Indicates the output upper limit of external wind power; Represents the historical forecast average value of external wind power output; εwo,t,q Represents the historical forecast error, which follows a normal distribution represents the standard deviation of historical forecast errors; Indicates the real-time forecast average; Represents the real-time prediction error, which also conforms to the mean of 0; represents the normal distribution of standard deviation, N() represents the normal distribution; w represents the number of wind turbines in the wind storage; and A 0-1 variable indicating the upper and lower limits of the output of the internal wind turbines have been reached. 0 means the internal fan output has not reached the upper limit; 1 means the internal fan output has reached the upper limit. 0 means that the internal fan output has not reached the lower limit; 1 means the internal fan output has reached the lower limit; and Respectively represent the historical forecast average value of internal wind power output, the historical output upper limit and the historical output lower limit; Γ 1 , Γ 2 They represent the conservative coefficients of the wind turbine number dimension and the time dimension respectively, e represents the number of energy storage; T represents the number of the last time period; and Respectively represent the real-time charging power of the internal energy storage and the real-time discharging power of the internal energy storage; and They represent the maximum value of charging and discharging of energy storage respectively; An indicator variable representing charging and discharging; represents the stored energy at time t; represents the stored energy at time t-1; and Respectively represent the upper and lower limits of the stored energy; η ec and η ed Respectively represent the charging efficiency and discharging efficiency of energy storage; E e,0 represents the initial value of stored energy; Indicates the final value of the stored energy in the real-time state; and They represent the output of internal wind power, the charging power of energy storage, and the discharging power of energy storage in the day-ahead state respectively; Indicates the maximum output value of internal wind power; The energy stored in the energy storage representing the state of the day before; Indicates the energy stored at the end of the energy storage in the day-ahead state; The energy stored in the previous time period represents the energy stored in the energy storage of the day-ahead state; With Oacmax They represent the charging energy consumption cost and the upper limit of the charging energy consumption cost of the wind storage system respectively; With O admax They represent the discharge energy consumption cost and the upper limit of the discharge energy consumption cost of the wind storage system respectively; and They represent the charging and discharging capacity of the wind storage system respectively. It is the state variable of charging and discharging of the wind storage system.

[0068] S2. Determine the clearing objective function and establish a clearing model based on factors other than the wind storage system.

[0069] S3. A double-layer model of the wind-storage system is established according to the energy consumption model obtained in step S1 and the clearing model obtained in step S2. The energy consumption model obtained in step S1 is the upper model, and the clearing model obtained in step S2 is the lower model.

[0070] The clearing objective function in step S2 is specifically set according to the following formula:

[0071]

[0072] Among them, g represents the number of ordinary units; G represents the number of ordinary units; They represent the energy consumption costs of traditional units, loads and independent wind power stations respectively; Indicates the power output of ordinary units; Indicates the load clearing amount; It indicates the power cleared by external independent wind power on the previous day.

[0073] The established clearing model satisfies the following constraints:

[0074]

[0075] Among them, the wind storage system is connected to the na node of the main grid; d is the load number; Indicates the output upper limit of ordinary units; and denote the dual variables of the upper and lower constraints of the traditional unit respectively; and They represent the upper and lower limits of load clearing respectively; and The dual variables representing the upper and lower constraints of the loading respectively; and The upper and lower limits of wind-storage system charging and discharging and the dual variables of upper and lower limit constraints are represented respectively; represents the output of independent wind power plants; represents the output upper limit of external wind power, and are the dual variables representing the upper and lower limit constraints of the wind turbine output respectively; represents the phase angle of node n in the system, represents the phase angle of node m in the system; Indicates the transmission upper limit of branch n to m; B nm Represents the nodal admittance matrix of the system; The dual variable representing the transmission upper limit constraint of transmission line n to m; and are the dual variables representing the upper and lower bound constraints on the phase angle of node n; represents the phase angle of the reference node; The dual variable representing the reference node phase angle constraint; d∈Ψ n represents the node to which the load is connected; g∈Ψ n Indicates the node connected to the common unit; a∈Ψ n represents the node connected to the wind storage system; w∈Ψ n Represents the node to which the external independent wind power is connected; m∈Θ n Indicates that m is a node connected to n; It represents the clearing unit energy consumption value of the power system at the day-ahead node.

[0076] S4. The built-in Carlo-Kuhn-Tucker condition method and the improved constraint and column generation algorithm are used in turn to optimize and solve the two-layer model of the wind-storage system established in step S3 to obtain the minimum total energy consumption of the wind-storage system.

[0077] S41. Optimize the double-layer model of the wind-storage system established in step S3 into a single-layer model using the embedded Carlo-Kuhn-Tucker condition method.

[0078] The objective function of the single-layer model obtained in step S41 is specifically set according to the following formula:

[0079]

[0080] in, Indicates the real-time energy consumption cost; represents the day-ahead energy cost.

[0081] The single-layer model obtained in step S41 satisfies the following constraints:

[0082]

[0083]

[0084] Where n(m∈Θ) represents the node number; represents the day-ahead cleared output power of external independent wind power; Indicates the real-time output power of internal wind power; Indicates the clearing of the connection nodes of the common units on the previous day; It represents the clearing unit energy consumption value of the day-ahead load connection node; It represents the clearing unit energy consumption value of the node connected to the wind storage system on the previous day; It represents the clearing unit energy consumption value of the external independent wind power connection node on the day before; The day-ahead node clearing unit energy consumption value of a node in a branch of the power system; It represents the unit energy consumption value of another node in a branch of the power system cleared the day before; Indicates the unit energy consumption value reported by ordinary units; Indicates the reported unit energy consumption value of the load; Indicates the reported unit energy consumption value of the wind storage system charging; Indicates the reported unit energy consumption value of the wind storage system discharge; Indicates the reported unit energy consumption value of the external independent fan; Indicates the clearing load value of the load under the day-ahead state. S42, using the improved constraint and column generation algorithm to solve the single-layer model obtained in step S41;

[0085] 1) Divide the objective function of the single-layer model obtained in step S41 into the main problem and sub-problems And add constraints The η q Represents a minimum total energy consumption parameter set; Represents the real-time k-th energy consumption cost.

[0086] 2) Set the upper bound of the objective function value to LB = -∞ and the lower bound to UB = ∞, and initialize the number of iterations k, the convergence limit parameter v, and the robust uncertainty parameter Preset the convergence limit parameter v.

[0087] 3) Use the solver to solve the main problem Solve and obtain the optimal day-ahead energy cost of the main problem and the optimal minimum total energy consumption parameter set and use the best day-ahead energy cost and the optimal minimum total energy consumption parameter set Update the lower bound to

[0088] 4) The optimal minimum total energy consumption parameter set of the main problem obtained in step 3) Substitute into the subproblem The solver is used to solve the optimal real-time energy consumption cost and utilize the best real-time energy cost Update the upper bound to

[0089] 5) According to the updated upper and lower bounds, determine Whether it meets:

[0090] If satisfied, stop the iteration and get the solution;

[0091] If not satisfied, repeat steps 3) to 5) and update the number of iterations k.

[0092] The innovation of the present invention is that it fully considers the uncertainty factors inside and outside the wind storage system, the internal component factors and the factors outside the wind storage system, and adopts the embedded Carlo-Kuhn-Tucker condition method and the improved constraint and column generation algorithm to achieve comprehensive consideration and full coverage of factors, so that the operating parameter evaluation is more in line with reality, and the minimum energy consumption obtained can bring the maximum power storage value.

Claims

1. A robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties, characterized in that: include: S1. Determine the energy consumption objective function of the wind storage system and establish the energy consumption model of the wind storage system according to the uncertain factors and internal components of the wind storage system; S2. Determine the clearing objective function and establish a clearing model based on factors other than the wind storage system; S3, establishing a double-layer model of the wind-storage system according to the energy consumption model obtained in step S1 and the clearing model obtained in step S2; S4. The built-in Carlo-Kuhn-Tucker condition method and the improved constraint and column generation algorithm are used in turn to optimize and solve the two-layer model of the wind-storage system established in step S3 to obtain the minimum total energy consumption of the wind-storage system.

2. A robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 1, characterized in that: The step S1 is specifically as follows: S11, modeling the uncertain factors outside the wind storage system, wherein the uncertain factors outside the wind storage system include output fluctuation and real-time energy consumption cost outside the wind storage system; S12, modeling the uncertain factors within the wind storage system, wherein the uncertain factors within the wind storage system include output fluctuations within the wind storage system; S13, modeling the internal components of the wind storage system, wherein the internal components of the wind storage system include an energy storage system and a wind turbine generator set of the wind storage system; S14. According to the modeling of steps S11 to S13, determine the energy consumption target function of the wind-storage system and establish an energy consumption model of the wind-storage system.

3. The robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 1 is characterized in that: The energy consumption objective function of the wind energy storage system in step S1 is specifically set according to the following formula: Among them, s represents the real-time scenario; t represents the time point; q represents the scenario where the day-ahead uncertainty is generated; w represents the number of the wind turbine in the wind storage; ρ q and ρ s represents the probability of a single scenario generated by two scenarios; represents the unit energy consumption cost of wind turbine abandonment, represents the clearing energy consumption cost of the node where the wind storage system is located on the day before; and They represent the day-ahead charging power of the wind-storage system, the day-ahead discharging power of the wind-storage system, the real-time output of internal wind power, and the difference between the real-time and day-ahead power of the wind-storage system; Indicates the actual predicted value of the real-time energy consumption cost scenario; Indicates the actual output upper limit of internal wind power; It represents the real-time clearing output value of internal wind power; maxmin represents the maximum and minimum value problem, that is, solving the optimal state under the worst condition.

4. The robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 1 is characterized in that: The energy consumption model of the wind energy storage system in step S1 satisfies the following constraints: Among them, wo and na represent the code of the external wind turbine and the node number of the wind storage connection respectively; Indicates the output upper limit of external wind power; Represents the historical forecast average value of external wind power output; ε wo,t,q represents the historical forecast error; represents the standard deviation of historical forecast errors; Represents the real-time forecast average; Represents the real-time prediction error; represents the normal distribution of standard deviation, N() represents the normal distribution; w represents the number of wind turbines in the wind storage; and A 0-1 variable indicating the upper and lower limits of the output of the internal wind turbines have been reached. 0 means the internal fan output has not reached the upper limit; 1 means the internal fan output has reached the upper limit. 0 means that the internal fan output has not reached the lower limit; 1 means the internal fan output has reached the lower limit; and They represent the historical forecast average value, historical output upper limit and historical output lower limit of internal wind power output respectively; Γ1, Γ2 represent the conservative coefficients of wind turbine number dimension and time dimension respectively, e represents the number of energy storage; T represents the number of the last time period; and Respectively represent the real-time charging power of the internal energy storage and the real-time discharging power of the internal energy storage; and They represent the maximum value of charging and discharging of energy storage respectively; An indicator variable representing charging and discharging; represents the stored energy at time t; represents the stored energy at time t-1; and Respectively represent the upper and lower limits of the stored energy; η ec and η ed Respectively represent the charging efficiency and discharging efficiency of energy storage; E e,0 represents the initial value of stored energy; Indicates the final value of the stored energy in the real-time state; and They represent the output of internal wind power, the charging power of energy storage, and the discharging power of energy storage in the day-ahead state respectively; Indicates the maximum output of internal wind power; The energy stored in the energy storage representing the state of the day before; Indicates the energy stored at the end of the energy storage in the day-ahead state; The energy stored in the previous time period represents the energy stored in the energy storage of the day-ahead state; With O acmax They represent the charging energy consumption cost and the upper limit of the charging energy consumption cost of the wind storage system respectively; With O admax They represent the discharge energy consumption cost and the upper limit of the discharge energy consumption cost of the wind storage system respectively; and They represent the charging and discharging capacity of the wind storage system respectively. It is the state variable of charging and discharging of the wind storage system.

5. The robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 1 is characterized in that: The clearing objective function in step S2 is specifically set according to the following formula: Among them, g represents the number of ordinary units; G represents the number of ordinary units; They represent the energy consumption costs of traditional units, loads and independent wind power stations respectively; Indicates the power output of ordinary units; Indicates the load clearing amount; It indicates the power output of external independent wind power on the previous day.

6. The robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 1 is characterized in that: The clearing model established in step S2 satisfies the following constraints: Where, d is the load number; Indicates the output upper limit of ordinary units; and denote the dual variables of the upper and lower constraints of the traditional unit respectively; and They represent the upper and lower limits of load clearing respectively; and The dual variables representing the upper and lower constraints of the loading respectively; and The upper and lower limits of wind-storage system charging and discharging and the dual variables of upper and lower constraints are represented respectively; represents the output of independent wind power plants; represents the output upper limit of external wind power, and are the dual variables representing the upper and lower limit constraints of the wind turbine output respectively; represents the phase angle of node n in the system, represents the phase angle of node m in the system; Indicates the transmission upper limit of branch n to m; B nm Represents the nodal admittance matrix of the system; The dual variable representing the transmission upper limit constraint of transmission line n to m; and are the dual variables representing the upper and lower bound constraints on the phase angle of node n; represents the phase angle of the reference node; The dual variable representing the reference node phase angle constraint; d∈Ψ n represents the node to which the load is connected; g∈Ψ n Indicates the node connected to the common unit; a∈Ψ n represents the node connected to the wind storage system; w∈Ψ n Represents the node to which the external independent wind power is connected; m∈Θ n Indicates that m is a node connected to n; It represents the clearing unit energy consumption value of the power system at the day-ahead node.

7. The robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 1 is characterized in that: The step S4 is specifically as follows: S41, optimizing the double-layer model of the wind-storage system established in step S3 into a single-layer model using an embedded Carlo-Kuhn-Tucker condition method; S42, solving the single-layer model obtained in step S41 using an improved constraint and column generation algorithm; S43. Obtain the minimum total energy consumption of the wind storage system based on the solution results.

8. The robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 7 is characterized in that: The objective function of the single-layer model obtained in step S41 is specifically set according to the following formula: in, Indicates the real-time energy consumption cost; represents the day-ahead energy cost.

9. The robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 7 is characterized in that: The single-layer model obtained in step S41 satisfies the following constraints: Where n(m∈Θ) represents the node number; represents the day-ahead cleared output power of external independent wind power; Indicates the real-time output power of internal wind power; Indicates the clearing of the connection nodes of the common units on the previous day; It represents the clearing unit energy consumption value of the day-ahead load connection node; It represents the clearing unit energy consumption value of the node connected to the wind storage system on the previous day; It represents the clearing unit energy consumption value of the external independent wind power connection node on the day before; The day-ahead node clearing unit energy consumption value of a node in a branch of the power system; It represents the unit energy consumption value of another node in a branch of the power system cleared the day before; Indicates the unit energy consumption value reported by ordinary units; Indicates the reported unit energy consumption value of the load; Indicates the reported unit energy consumption value of the wind storage system charging; Indicates the reported unit energy consumption value of the wind storage system discharge; Indicates the reported unit energy consumption value of the external independent fan; Indicates the clearing load value of the load in the day-ahead state.

10. The robust-stochastic decision control method for a wind-storage system taking into account internal and external uncertainties according to claim 8, characterized in that: The step S42 is specifically as follows: 1) Divide the objective function of the single-layer model obtained in step S41 into the main problem and sub-problems And add constraints The η q Represents a minimum total energy consumption parameter set; represents the real-time k-th energy consumption cost; 2) Set the upper bound of the objective function value to LB = -∞ and the lower bound to UB = ∞, and initialize the number of iterations k, the convergence limit parameter v, and the robust uncertainty parameter Preset convergence limit parameter v; 3) Use the solver to solve the main problem Solve and obtain the optimal day-ahead energy cost of the main problem and the optimal minimum total energy consumption parameter set and use the best day-ahead energy cost and the optimal minimum total energy consumption parameter set Update the lower bound to 4) The optimal minimum total energy consumption parameter set of the main problem obtained in step 3) Substitute into the subproblem The solver is used to solve the optimal real-time energy consumption cost and utilize the best real-time energy cost Update the upper bound to 5) According to the updated upper and lower bounds, determine Whether it meets: If satisfied, stop the iteration and get the solution; If not satisfied, repeat steps 3) to 5) and update the number of iterations k.

Images