Distribution robust chance-constrained power system day-ahead dispatch method based on adaptive reinforced linear approximation

Abstract

The application discloses a kind of distribution robust chance constraint power system day-ahead scheduling methods based on adaptive reinforcement linear approximation, comprising:1, obtains new energy historical output prediction error sample set and constructs wasserstein distribution fuzzy set;2, establish distribution robust joint chance constraint power system day-ahead scheduling model;3, to distribution robust joint chance constraint is reinforced linear approximation reconstruction, establishes linear optimization model containing sample scaling factor;4, obtains initial scheduling scheme using soft constraint cold start strategy;5, based on constraint violation depth adaptive update sample scaling factor, and iteratively solve linear optimization model, obtain power system day-ahead scheduling scheme.The application can realize the adaptive distribution of new energy prediction error sample robust margin, to improve the feasibility of distribution robust chance constraint model under extreme prediction error scenario, so as to improve the safety and economy of power system day-ahead scheduling under the condition of high proportion of new energy access.

Description

Technical Field

[0001] This invention belongs to the field of power system dispatching technology, specifically relating to a day-ahead dispatching method for power systems based on adaptive reinforced linear approximation and bibliometric chance constraints. Background Technology

[0002] With the continuous growth of installed capacity of new energy sources, the output of new energy sources such as wind power and photovoltaics exhibits strong randomness and volatility, making day-ahead dispatching of the power system face more significant uncertainties. Traditional deterministic day-ahead dispatching methods cannot fully account for the operational risks caused by new energy forecasting errors, which can easily lead to problems such as insufficient reserves and line overruns. Therefore, it is necessary to establish a day-ahead dispatching method that can take into account the uncertainties of new energy sources in order to improve the safety and economy of power system operation.

[0003] The robust chance constraint method uses Wasserstein distributed fuzzy sets to describe the probability distribution uncertainty of renewable energy prediction errors, effectively controlling system operation risks under conditions where the true probability distribution is unknown. However, existing reinforced linear approximation methods typically employ fixed sample scaling strategies, which fail to reflect the actual impact of different prediction error samples on operation risks. This results in extreme prediction error samples occupying excessive robustness margins, reducing model feasibility and hindering the safe and stable operation of day-ahead dispatching of the power system under conditions of high renewable energy integration. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by proposing a day-ahead scheduling method for power systems based on adaptive reinforced linear approximation and a sub-Blu-ray opportunity constraint. The aim is to achieve adaptive allocation of robust margins for renewable energy prediction error samples, thereby improving the feasibility of the sub-Blu-ray opportunity constraint model under extreme prediction error scenarios. This, in turn, enhances the safety, economy, and renewable energy absorption capacity of the power system's day-ahead scheduling under conditions of high renewable energy integration.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: The present invention provides a day-ahead scheduling method for a power system based on an adaptive reinforced linear approximation and a distributed bar chance constraint, characterized by the following steps: Step 1: Obtain a sample set of historical power output prediction errors for new energy sources Used to construct Wasserstein distributed fuzzy sets ;in, The first error in the historical power output prediction of new energy One sample, The total number of samples, Represents the Wasserstein radius; Step 2: Establish a day-ahead dispatch model for the power system with joint chance constraints and distributed bar structure, including: an objective function aimed at minimizing generator coal consumption within the dispatch period, deterministic operating constraints, and a Wasserstein-based fuzzy set model. Constructed joint opportunity constraints for the sub-Brussels bar; Step 3: According to The joint chance constraint of the split-bar is reinforced by a linear approximation reconstruction to construct tail sample constraints, robust margin constraints and non-tail sample support constraints. After relaxing the robust margin constraints, the relaxed robust margin constraints are obtained. Thus, the objective function, deterministic running constraints, tail sample constraints, relaxed robust margin constraints and non-tail sample support constraints constitute a linear optimization model with a sample scaling factor. Step 4: Solve the linear optimization model with sample scaling factor to obtain the final day-ahead dispatch scheme of the power system, that is, the optimal planned power of each generator under a single dispatch cycle.

[0006] The method of the present invention is also characterized in that, in step one, an empirical probability distribution is established using equation (1). ,by Centered on, a Wasserstein distributed fuzzy set is constructed. ,in, express The probability distribution contained therein express arrive Wasserstein distance; (1) In equation (1), Indicated by Dirac measure for support points.

[0007] Furthermore, step two includes: Step 2.1: Use equation (2) to establish the objective function of the day-ahead dispatch model of the power system. : (2) In equation (2), This represents the total number of time periods within a single scheduling cycle. Indicates the number of generators. Indicates the first The coal consumption coefficient of each generator. Indicates the first The first time period The planned power of each generator; Step 2.2: Establish deterministic operating constraints for the day-ahead dispatch model of the power system; Step 2.2.1: Construct power balance constraints using equation (3); (3) In equation (3), Indicates the number of new energy sources. Indicates the first The first time period The predicted output of new energy sources Indicates the number of loads. Indicates the first The first time period The power required by each load; Step 2.2.2: Construct the operating range constraints of the generator using equation (4); (4) In equation (4), and They represent the first The upper and lower power limits of each generator; Step 2.2.3: Construct the ramp constraint for the generator using equation (5); (5) In equation (5), Indicates the first The first time period The planned power of each generator and They represent the first The upper and lower limits of the ramp power of each generator; Step 2.3, based on Using equation (6), we establish the joint chance constraint of the sub-Brukers: (6) In equation (6), supremacy represents the supremacy. Represents probability. This represents the vector of decision variables in the day-ahead dispatch model of the power system. A vector of random variables representing the prediction error of new energy sources. This represents the set of safety constraints in the combined opportunity constraints of the sub-Bruker rod, including: generator operating range constraints and line power flow safety constraints; Indicates transpose. Indicates safety constraints The corresponding decision variable coefficient vector, and , Indicates safety constraints The corresponding random variable coefficient vector, Indicates safety constraints The corresponding constraint boundaries, This indicates the upper limit of the allowed system risk probability.

[0008] Furthermore, step three includes: Step 3.1: Calculate the first step using equation (7). Each sample under safety constraints Projection value in direction And based on the projection values, for Sort in ascending order to select the top sorted items. A number of samples constitute a security constraint. The corresponding tail sample set and define security constraints Corresponding quantile boundaries ;in, Indicates the sorted order of the first... Each sample under safety constraints Projection value in the direction; (7) Step 3.2: Construct robust margin constraints using equation (8): (8) In equation (8), Denotes the robustness margin variable, and , Indicates the first The slack variables corresponding to each sample, and ; Step 3.3, based on Tail sample constraints are constructed using equation (9): (9) In equation (9), express The Middle Each tail sample is under safety constraints Projection value in the direction, express The dual norm, express The Middle Scaling factor corresponding to each tail sample; Step 3.4, based on Using equation (10), non-tail sample support constraints are constructed: (10) Step 3.5: Construct the relaxed robust margin constraint using equation (11): (11) In equation (11), These are auxiliary slack variables.

[0009] Furthermore, step four includes: Step 4.1: Define the current iteration number as... and initialize ; Define and initialize the first In the nth iteration The sample scaling factor corresponding to each tail sample ; based on , construct the first Tail sample constraints in the nth iteration, and for those containing the nth iteration Solving the linear optimization model with tail sample constraints in the nth iteration yields the nth iteration. Decision variables in the next iteration ; Step 4.2, based on Calculate the constraint violation depth corresponding to each tail sample, which is used to update the first tail sample. The scaling factor for each tail sample in the nth iteration is obtained. Scaling factor for each tail sample in each iteration; Step 4.3, based on the first The scaling factor for each tail sample in the nth iteration is then removed, and the auxiliary slack variables are removed to obtain the nth iteration. The tail sample constraints in the nth iteration, together with the objective function, deterministic running constraints, robustness margin constraints, and non-tail sample support constraints, constitute the reconstructed reinforced linear approximate optimization model, which is then solved to obtain the nth iteration. Decision variables for the next iteration ; Step 4.4, if Then output As the final day-ahead dispatching scheme for the power system, otherwise, Assigned Return to step 4.2 and continue execution, where, Indicates the convergence threshold. It represents the 2-norm.

[0010] Furthermore, step 4.2 includes: Step 4.2.1, based on Calculate the first using equation (12) In the next iteration The Middle The constraint violation depth corresponds to each tail sample. : (12) Step 4.2.2, based on Calculate the first using equation (13) In the next iteration The Middle Scaling factor for each tail sample : (13) In equation (13), The selected constraint violates the depth threshold.

[0011] Furthermore, in step 4.3, based on the first The iteration of the iteration Tail sample scaling factor Construct the first according to formula (14) Tail sample constraints in the next iteration: (14).

[0012] The present invention provides an electronic device, including a memory and a processor, characterized in that the memory is used to store a program that supports the processor in executing a day-ahead scheduling method for a sub-Bruker chance-constrained power system based on an adaptive reinforced linear approximation, and the processor is configured to execute the program stored in the memory.

[0013] The present invention discloses a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, performs the steps of a day-ahead scheduling method for a distributed bar chance constraint power system based on an adaptive reinforced linear approximation.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention dynamically identifies high-risk prediction error scenarios by adaptively adjusting the sample scaling factor based on the constraint violation depth of new energy prediction error samples. It rationally allocates the robustness margin of different prediction error samples, avoiding excessive consumption of margin by extreme samples that leads to conservative or infeasible models. This improves the feasibility of day-ahead scheduling models under extreme new energy prediction error scenarios and enhances the safe operation capability of power systems in the face of severe uncertainty events.

[0015] 2. This invention continuously improves the day-ahead dispatch scheme of the power system through multiple rounds of iterative optimization. In each iteration, the feasible region is adaptively tightened according to the constraint violation situation, gradually approaching the true optimal solution of the distributed bar chance constraint. Under the premise of ensuring that the system operation risk is controllable, it can reduce the coal consumption of generators, reduce unnecessary reserve reserves, and improve the level of new energy consumption, thereby achieving an overall improvement in the economic efficiency of power system operation. Attached Figure Description

[0016] Figure 1This is a flowchart of the day-ahead scheduling method for a power system with distributed bar chance constraints according to the present invention. Detailed Implementation

[0017] In this embodiment, a day-ahead scheduling method for a sub-Bruker chance-constrained power system based on adaptive reinforced linear approximation is described, such as... Figure 1 As shown, it includes the following steps: Step 1: Obtain a sample set of historical power output prediction errors for new energy sources ,in, The first error in the historical power output prediction of new energy One sample, The total number of samples; Establish the empirical probability distribution using equation (1) ,by Centered on, a Wasserstein distributed fuzzy set is constructed. ,in, express The probability distribution contained therein express arrive Wasserstein distance.

[0018] (1) In equation (1), Indicated by Dirac measure for support points.

[0019] Historical samples of new energy prediction errors can reflect the uncertainty characteristics of new energy output. By using historical samples to construct an empirical probability distribution and further establishing a Wasserstein distribution fuzzy set, the uncertainty of the new energy prediction error can be described without assuming that it follows a specific probability distribution, thereby improving the adaptability of the subsequent split-bar joint chance constraint model to the distribution shift of the prediction error.

[0020] Step 2: Establish a day-ahead dispatch model for the power system with joint chance constraints and distributed bar structure, including: an objective function aimed at minimizing generator coal consumption within the dispatch period, deterministic operating constraints, and a Wasserstein-based fuzzy set model. A combined opportunity constraint of the distributed Brussels rods was constructed. This day-ahead dispatch model of the power system is used to comprehensively consider the operational risks caused by generator coal consumption, unit operation constraints, and new energy prediction errors, and determines the day-ahead output plan of each generator unit under the premise of meeting the system's safe operation requirements.

[0021] Step 2.1: Use equation (2) to establish the objective function of the day-ahead dispatch model of the power system. : (2) In equation (2), This represents the total number of time periods within a single scheduling cycle. Indicates the number of generators. Indicates the first The coal consumption coefficient of each generator. Indicates the first The first time period The planned power of each generator.

[0022] Step 2.2: Establish deterministic operating constraints for the day-ahead dispatch model of the power system.

[0023] Step 2.2.1: Construct power balance constraints using equation (3): (3) In equation (3), Indicates the number of new energy sources. Indicates the first The first time period The predicted output of new energy sources Indicates the number of loads. Indicates the first The first time period The power required by each load.

[0024] Step 2.2.2: Construct the generator's operating range constraints using equation (4): (4) In equation (4), and They represent the first The upper and lower power limits of each generator; Step 2.2.3: Construct the generator's ramp constraint using equation (5): (5) In equation (5), Indicates the first The first time period The planned power of each generator and They represent the first The upper limit and lower limit of the ramp power of each generator.

[0025] Step 2.3, based on Using equation (6), we establish the joint chance constraint of the sub-Brukers: (6) In equation (6), supremacy represents the supremacy. Represents probability. This represents the vector of decision variables in the day-ahead dispatch model of the power system. A vector of random variables representing the prediction error of new energy sources. This represents the set of safety constraints in the combined opportunity constraints of the sub-Bruker rod, including: generator operating range constraints and line power flow safety constraints; Indicates transpose. Indicates safety constraints The corresponding decision variable coefficient vector, and , Indicates safety constraints The corresponding random variable coefficient vector, Indicates safety constraints The corresponding constraint boundaries, This indicates the upper limit of the allowed system risk probability.

[0026] Step 3: According to The joint chance constraint of the split bar is reconstructed by strengthening the linear approximation to construct the tail sample constraint, robust margin constraint and non-tail sample support constraint. After relaxing the robust margin constraint, the relaxed robust margin constraint is obtained. Thus, the objective function, deterministic running constraint, tail sample constraint, relaxed robust margin constraint and non-tail sample support constraint constitute a linear optimization model with sample scaling factor.

[0027] Step 3.1: Calculate the first step using equation (7). Each sample under safety constraints Projection value in direction And based on the projection values, for Sort in ascending order to select the top sorted items. A number of samples constitute a security constraint. The corresponding tail sample set and define security constraints Corresponding quantile boundaries ;in, Indicates the sorted order of the first... Each sample under safety constraints Projection value in the direction.

[0028] (7) Since the system's operational risk is mainly determined by the tail samples with large prediction errors, the samples are sorted according to their projection values, and the tail sample set is extracted to focus on characterizing the high-risk prediction error samples that affect the probability of satisfying the chance constraints, thereby improving the ability of the enhanced linear approximation model to describe tail risks.

[0029] Step 3.2: Construct robust margin constraints using equation (8): (8) In equation (8), Denotes the robustness margin variable, and , Indicates the first The slack variables corresponding to each sample, and .

[0030] Among them, the robust margin variable is used to measure the degree of worst distribution shift caused by the fuzzy set of the Wasserstein distribution. By uniformly constraining the robust margin occupancy of each prediction error sample, the established linear optimization model can meet the requirements of the bisserstein joint chance constraint.

[0031] Step 3.3, based on Tail sample constraints are constructed using equation (9): (9) In equation (9), express The Middle Each tail sample is under safety constraints Projection value in the direction, express The dual norm, express The Middle The scaling factor corresponding to each tail sample.

[0032] Step 3.4, based on Using equation (10), non-tail sample support constraints are constructed: (10) Step 3.5: Construct the relaxed robust margin constraint using equation (11): (11) In equation (11), These are auxiliary slack variables.

[0033] In the first iteration, since the scaling factors of each tail sample have not yet been adaptively updated, directly solving the reinforced linear approximation model may easily lead to model infeasibility. Therefore, auxiliary slack variables are introduced to construct slack robust margin constraints, thereby obtaining initial decision variables that satisfy the constraints and providing an initial solution for the adaptive update of the subsequent sample scaling factors.

[0034] Step 4: Solve the linear optimization model with sample scaling factor to obtain the final day-ahead dispatch scheme of the power system, that is, the optimal planned power of each generator under a single dispatch cycle.

[0035] Step 4.1: Define the current iteration number as... and initialize .

[0036] Define and initialize the first In the nth iteration The sample scaling factor corresponding to each tail sample .

[0037] based on Construct the first according to formula (12) Tail sample constraints in the nth iteration, and for those containing the nth iteration Solving the linear optimization model with tail sample constraints in the nth iteration yields the nth iteration. Decision variables in the next iteration : (12) Step 4.2, based on Calculate the constraint violation depth corresponding to each tail sample, which is used to update the first tail sample. The scaling factor for each tail sample in the nth iteration is obtained. Scaling factor for each tail sample in each iteration.

[0038] Step 4.2.1, based on Calculate the first using equation (13) In the next iteration The Middle The constraint violation depth corresponds to each tail sample. : (13) The constraint violation depth is used to measure the degree of influence of each tail sample on the current decision scheme's violation of safety constraints. The greater the violation depth, the higher the contribution of the corresponding prediction error sample to the system's operational risk. Therefore, a larger sample scaling factor should be allocated in subsequent iterations to enhance the model's attention to high-risk samples.

[0039] Step 4.2.2, based on Calculate the first using equation (14) In the next iteration The Middle Scaling factor for each tail sample : (14) In equation (14), The selected constraint violation depth threshold limits the maximum value of the sample scaling factor. When the constraint violation depth corresponding to the tail sample is less than [a certain value], the maximum value is determined by the constraint violation depth threshold. When the constraint violation depth threshold is set, the scaling factor of the corresponding sample is updated according to equation (13); otherwise, the sample scaling factor is kept at 1. By setting the constraint violation depth threshold, the frequent adjustment of the sample scaling factor caused by slight constraint violations can be avoided, thereby improving the stability of the iterative solution process and ensuring effective correction of the tail samples with serious constraint violations.

[0040] Step 4.3, based on the first The iteration of the iteration Tail sample scaling factor And remove the auxiliary slack variables, and construct the first according to equation (15). The tail sample constraints of the second iteration, together with the objective function, deterministic running constraints, robustness margin constraints, and non-tail sample support constraints, constitute the reconstructed reinforced linear approximate optimization model, which is then solved to obtain the third iteration. Decision variables for the next iteration .

[0041] (15) By substituting the updated sample scaling factor back into the reinforced linear approximation model, the robust margin occupancy corresponding to each tail sample can be redistributed, allowing the model to gradually approach the true feasible region of the split-Bruker joint chance constraint, thereby further optimizing the day-ahead scheduling results.

[0042] Step 4.4, if Then output As the final day-ahead dispatching scheme for the power system, otherwise, Assigned Return to step 4.2 and continue execution, where, Indicates the convergence threshold. Denotes the 2-norm and the convergence threshold. Used to control the precision of iteration termination. When the change in decision variables between two consecutive iterations is less than the convergence threshold, the current scheduling scheme can be considered to have converged.

[0043] In this embodiment, an electronic device includes a memory and a processor. The memory stores a program that supports the processor in executing the above-described method, and the processor is configured to execute the program stored in the memory.

[0044] In this embodiment, a computer-readable storage medium stores a computer program, which is executed by a processor to perform the steps of the above method.

Claims

1. A day-ahead scheduling method for a distributed bar chance-constrained power system based on adaptive reinforced linear approximation, characterized in that, Includes the following steps: Step 1: Obtain a sample set of historical power output prediction errors for new energy sources Used to construct Wasserstein distributed fuzzy sets ;in, The first error in the historical power output prediction of new energy One sample, The total number of samples, Represents the Wasserstein radius; Step 2: Establish a day-ahead dispatch model for the power system with joint chance constraints and distributed bar structure, including: an objective function aimed at minimizing generator coal consumption within the dispatch period, deterministic operating constraints, and a Wasserstein-based fuzzy set model. Constructed joint opportunity constraints for the sub-Brussels bar; Step 3: According to The joint chance constraint of the split-bar is reinforced by a linear approximation reconstruction to construct tail sample constraints, robust margin constraints and non-tail sample support constraints. After relaxing the robust margin constraints, the relaxed robust margin constraints are obtained. Thus, the objective function, deterministic running constraints, tail sample constraints, relaxed robust margin constraints and non-tail sample support constraints constitute a linear optimization model with a sample scaling factor. Step 4: Solve the linear optimization model with sample scaling factor to obtain the final day-ahead dispatch scheme of the power system, that is, the optimal planned power of each generator under a single dispatch cycle.

2. The method as claimed in claim 1, characterized in that, Step one involves using equation (1) to establish an empirical probability distribution. ,by Centered on, a Wasserstein distributed fuzzy set is constructed. ,in, express The probability distribution contained therein express arrive Wasserstein distance; (1) In equation (1), Indicated by Dirac measure for support points.

3. The method as described in claim 1, characterized in that, Step two includes: Step 2.1: Use equation (2) to establish the objective function of the day-ahead dispatch model of the power system. : (2) In equation (2), This represents the total number of time periods within a single scheduling cycle. Indicates the number of generators. Indicates the first The coal consumption coefficient of each generator. Indicates the first The first time period The planned power of each generator; Step 2.2: Establish deterministic operating constraints for the day-ahead dispatch model of the power system; Step 2.2.1: Construct power balance constraints using equation (3); (3) In equation (3), Indicates the number of new energy sources. Indicates the first The first time period The predicted output of new energy sources Indicates the number of loads. Indicates the first The first time period The power required by each load; Step 2.2.2: Construct the operating range constraints of the generator using equation (4); (4) In equation (4), and They represent the first The upper and lower power limits of each generator; Step 2.2.3: Construct the ramp constraint for the generator using equation (5); (5) In equation (5), Indicates the first The first time period The planned power of each generator and They represent the first The upper and lower limits of the ramp power of each generator; Step 2.3, based on Using equation (6), we establish the joint chance constraint of the sub-Brukers: (6) In equation (6), supremacy represents the supremacy. Represents probability. This represents the vector of decision variables in the day-ahead dispatch model of the power system. A vector of random variables representing the prediction error of new energy sources. This represents the set of safety constraints in the combined opportunity constraints of the sub-Bruker rod, including: generator operating range constraints and line power flow safety constraints; Indicates transpose. Indicates safety constraints The corresponding decision variable coefficient vector, and , Indicates safety constraints The corresponding random variable coefficient vector, Indicates safety constraints The corresponding constraint boundaries, This indicates the upper limit of the allowed system risk probability.

4. The method as described in claim 1, characterized in that, Step three includes: Step 3.1: Calculate the first step using equation (7). Each sample under safety constraints Projection value in direction And based on the projection values, for Sort in ascending order to select the top sorted items. A number of samples constitute a security constraint. The corresponding tail sample set and define security constraints Corresponding quantile boundaries ;in, Indicates the sorted order of the first... Each sample under safety constraints Projection value in the direction; (7) Step 3.2: Construct robust margin constraints using equation (8): (8) In equation (8), Denotes the robustness margin variable, and , Indicates the first The slack variables corresponding to each sample, and ; Step 3.3, based on Tail sample constraints are constructed using equation (9): (9) In equation (9), express The Middle Each tail sample is under safety constraints Projection value in the direction, express The dual norm, express The Middle Scaling factor corresponding to each tail sample; Step 3.4, based on Using equation (10), non-tail sample support constraints are constructed: (10) Step 3.5: Construct the relaxed robust margin constraint using equation (11): (11) In equation (11), These are auxiliary slack variables.

5. The method as described in claim 1, characterized in that, Step four includes: Step 4.1: Define the current iteration number as... and initialize ; Define and initialize the first In the nth iteration The sample scaling factor corresponding to each tail sample ; based on , construct the first Tail sample constraints in the nth iteration, and for those containing the nth iteration Solving the linear optimization model with tail sample constraints in the nth iteration yields the nth iteration. Decision variables in the next iteration ; Step 4.2, based on Calculate the constraint violation depth corresponding to each tail sample, which is used to update the first tail sample. The scaling factor for each tail sample in the nth iteration is obtained. Scaling factor for each tail sample in each iteration; Step 4.3, based on the first The scaling factor for each tail sample in the nth iteration is then removed, and the auxiliary slack variables are removed to obtain the nth iteration. The tail sample constraints in the nth iteration, together with the objective function, deterministic running constraints, robustness margin constraints, and non-tail sample support constraints, constitute the reconstructed reinforced linear approximate optimization model, which is then solved to obtain the nth iteration. Decision variables for the next iteration ; Step 4.4, if Then output As the final day-ahead dispatching scheme for the power system, otherwise, Assigned Return to step 4.2 and continue execution, where, Indicates the convergence threshold. It represents the 2-norm.

6. The method as described in claim 1, characterized in that, Step 4.2 includes: Step 4.2.1, based on Calculate the first using equation (12) In the next iteration The Middle The constraint violation depth corresponds to each tail sample. : (12) Step 4.2.2, based on Calculate the first using equation (13) In the next iteration The Middle Scaling factor for each tail sample : (13) In equation (13), The selected constraint violates the depth threshold.

7. The method as described in claim 1, characterized in that, In step 4.3, based on the first The iteration of the iteration Tail sample scaling factor Construct the first according to formula (14) Tail sample constraints in the next iteration: (14)。 8. An electronic device, comprising a memory and a processor, characterized in that, The memory is used to store programs that support the processor in executing any of the day-ahead scheduling methods for a distributed bar chance-constrained power system based on adaptive reinforced linear approximation as described in claims 1-7, and the processor is configured to execute the programs stored in the memory.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When a computer program is run by a processor, it executes the steps of any of the day-ahead scheduling methods for a distributed bar chance-constrained power system based on adaptive reinforced linear approximation as described in claims 1-7.

Images