Micro-grid optimal scheduling method and system considering power grid peak regulation

Abstract

The invention discloses a micro-grid optimal scheduling method considering power grid peak regulation. The method comprises the following steps: acquiring data information of a target micro-grid and a power system; constructing a day-ahead new energy power scene set of the target microgrid; constructing a first-stage load base line making model of optimal scheduling of the target micro-grid and solving the first-stage load base line making model; constructing a second-stage declaration capacity decision-making model of optimal scheduling of the target microgrid and solving the second-stage declaration capacity decision-making model; and according to an obtained result, completing optimal scheduling of the target micro-grid in consideration of power grid peak regulation. The invention also discloses a system for realizing the micro-grid optimal scheduling method considering the peak regulation of the power grid. According to the method, the optimal scheduling process of the micro-grid considering the peak regulation basis of the power grid is realized, and the method is higher in reliability and better in accuracy.

Description

Technical Field

[0001] This invention belongs to the field of electrical automation, and specifically relates to a microgrid optimization scheduling method and system that takes into account power grid peak shaving. Background Technology

[0002] With economic and technological development and the improvement of people's living standards, electricity has become an indispensable secondary energy source in people's production and daily life, bringing endless convenience. Therefore, ensuring a stable and reliable supply of electricity has become one of the most important tasks of the power system.

[0003] As an emerging component of the power system, microgrids integrate distributed wind and solar power, energy storage, and load resources to achieve considerable regulation capacity and response speed, thus effectively responding to the grid's peak-shaving needs. In the current process of power system optimization and dispatching, the optimized dispatching of microgrids has become a crucial part.

[0004] However, current microgrid optimization dispatch schemes often fail to consider the overall power system dispatch process. For example, existing dispatch schemes (such as those mentioned in CN120184969A and CN120200321A) focus only on the microgrid's own dispatch process, neglecting the grid's peak-shaving needs and failing to design corresponding optimization strategies for peak-shaving assessment mechanisms. This results in relatively poor reliability and accuracy of the microgrid's optimization dispatch process, and it is no longer suitable for the current power system's operational requirements. Summary of the Invention

[0005] One of the objectives of this invention is to provide a highly reliable and accurate microgrid optimization scheduling method that takes into account grid peak shaving.

[0006] The second objective of this invention is to provide a system for implementing the microgrid optimal scheduling method that takes into account grid peak shaving.

[0007] The microgrid optimization scheduling method considering power grid peak shaving provided by this invention includes the following steps:

[0008] S1. Acquire data information about the target microgrid and power system;

[0009] S2. Based on the data obtained in step S1, construct a set of day-ahead renewable energy power scenarios for the target microgrid;

[0010] S3. Based on the scenario set constructed in step S2, construct and solve the first-stage load baseline formulation model for the optimal scheduling of the target microgrid;

[0011] S4. Based on the results obtained in step S3, construct and solve the second-stage capacity application decision model for the optimal scheduling of the target microgrid;

[0012] S5. Based on the results obtained in steps S3 and S4, complete the optimized scheduling of the target microgrid considering grid peak shaving.

[0013] Step S2, which involves constructing a day-ahead renewable energy power scenario set for the target microgrid based on the data information obtained in step S1, includes the following steps:

[0014] Obtain the predicted and actual data of each distributed renewable energy source under the target microgrid during each scheduling period;

[0015] Based on the Cornish-Fisher series, a cumulative probability function for the prediction error of renewable energy power in each scheduling period is constructed.

[0016] Based on the sorting method that takes into account time correlation, the prediction error scenarios for each scheduling period are obtained;

[0017] By overlaying the day-ahead renewable energy forecast data for each scheduling period with the forecast error scenario, the day-ahead renewable energy power scenario set for each period of the target microgrid is obtained.

[0018] Step S2 specifically includes the following steps:

[0019] Obtain the predicted and actual power data of each distributed renewable energy source under the target microgrid during each scheduling period; calculate the power prediction error:

[0020] In the formula Let be the power prediction error of the m-th new energy power station at time t on the j-th historical day; Let be the actual power output of the m-th new energy power station at time t on the j-th historical day; Let m be the predicted power of the m-th new energy power plant at time t on the j-th historical day; , For the collection of new energy power stations; , For time sets; , A collection of historical days;

[0021] Using a Cornish-Fisher series containing fifth-order cumulants, the cumulative probability function of the historical power prediction error for each renewable energy power plant is constructed:

[0022] In the formula Let be the first-order raw moment of the historical prediction error of the m-th new energy power station at time t; The total number of historical days; Let be the second-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the third-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the fourth-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the fifth-order original moment of the historical prediction error of the m-th new energy power station at time t; Let q be the quantile of the probability value q corresponding to the standardized prediction error CDF of the m-th new energy power station at time t. q represents the quantile of the probability value q corresponding to the standard Gaussian distribution; Let q be the quantile of the probability value q corresponding to the standardization prediction error (CDF) of the m-th new energy power station at time t.

[0023] Based on Latin hypercube sampling, the probability q in the cumulative probability function of the historical power prediction error for each new energy power station is set as follows: To obtain the cumulative probability function vector of the prediction error of new energy power in each time period;

[0024] The cumulative probability function vectors of the new energy power prediction errors obtained in each time period are concatenated to obtain the cumulative probability function vector matrix of errors. , represented as:

[0025] Will As the initial set of day-ahead prediction error scenarios;

[0026] right The permutation and combination of each row of elements in the matrix ensures that the final generated scene set matches the time correlation coefficient of the historical error matrix:

[0027] Calculations were performed on each new energy power station at any time. and Time correlation coefficient between historical error samples for ;

[0028] right The scene vector at each time step undergoes a set of corresponding elementary transformation matrices. Calculate and obtain the power prediction error scenario vector after sorting each time period. for:

[0029] The power prediction error scenario vector for all time periods is calculated using the following formula. The time correlation coefficient of the error scenario matrix:

[0030] In the formula For the m-th new energy power station and The time correlation coefficient of the error scenario matrix between time points; For the m-th new energy power station The power prediction error for the i-th scenario at time i; A collection of scenes; This represents the total number of scenes;

[0031] Optimization problem:

[0032] Solving the constructed optimization problem yields the elementary transformation matrix. This leads to the power prediction error scenario vector sorted by time period. ;

[0033] Finally, the power prediction error scenario vector obtained by superimposing the day-ahead power prediction values ​​of each renewable energy power plant is used to obtain the day-ahead renewable energy power scenario set, denoted as:

[0034] In the formula Let i be the power result of the m-th new energy power station at time t for the i-th scenario; The set that is composed of the current new energy power scenario set; Let t be the day-ahead predicted power of the m-th renewable energy power station at time t.

[0035] Step S3, based on the scenario set constructed in step S2, involves constructing and solving the first-stage load baseline determination model for the optimal scheduling of the target microgrid. Specifically, this includes the following steps:

[0036] Without considering microgrid peak shaving, a first-stage load baseline determination model for the optimal scheduling of the target microgrid is constructed and solved, with the total cost of the microgrid as the objective function and the energy storage constraints, power balance constraints, and tie-line constraints of the microgrid as constraints.

[0037] Step S3 specifically includes the following steps:

[0038] Objective function:

[0039] The following formula is used as the objective function for the first-stage load baseline determination model of the target microgrid's optimal scheduling:

[0040] In the formula Develop the objective function value for the model to determine the first-stage load baseline; The cost of purchasing and selling electricity from the main grid to the microgrid, and , Let be the electricity purchase price of the microgrid at time t. Let be the power purchased by the microgrid at time t. Let be the electricity price of the microgrid at time t. Let be the electricity sold by the microgrid at time t; The cost of energy storage charging and discharging in microgrids, and , A collection of energy storage power stations, Let be the charging and discharging cost coefficient of the e-th energy storage power station. Let e ​​be the charging power of the e-th energy storage power station at time t. Let e ​​be the charging efficiency of the e-th energy storage power station. Let be the discharge power of the e-th energy storage power station at time t. Let be the discharge efficiency of the e-th energy storage power station; The cost of curtailment in microgrids, and , Let i be the probability of the i-th scenario occurring. Let be the curtailment penalty cost coefficient for the m-th renewable energy power station. Let represent the power curtailment of the m-th renewable energy power station at time t in scenario i. Let be the load shedding penalty cost coefficient for the m-th renewable energy power station. Let be the load shedding power of the m-th new energy power station in scenario i at time t.

[0041] Constraints:

[0042] The following formula is used as the energy storage constraint for the microgrid:

[0043] In the formula Let e ​​be the minimum allowable discharge power of the e-th energy storage power station; Let be a binary variable representing the discharge state of the e-th energy storage power station at time t, and let the e-th energy storage power station be in a discharge state at time t. If the e-th energy storage power station is not in a discharging state at time t, then... ; This represents the maximum allowable discharge power of the e-th energy storage power station; The minimum allowed charging power for the e-th energy storage power station; Let be a binary variable representing the charging state of the e-th energy storage station at time t, and let be the charging state of the e-th energy storage station at time t. If the e-th energy storage station is not in a charging state at time t, then... ; This represents the maximum allowed charging power for the e-th energy storage power station. Let be the remaining power of the e-th energy storage power station at time t+1; The minimum allowable storage capacity for the e-th energy storage power station; The maximum allowable storage capacity for the e-th energy storage power station; The remaining power of the e-th energy storage station at the end of the day; The remaining power of the e-th energy storage station at the start of the day; Let be the depth of discharge coefficient of the e-th energy storage power station; Let be the installed capacity of the e-th energy storage power station;

[0044] The following formula is used as the power balance constraint for the microgrid:

[0045] In the formula Let be the power of the m-th renewable energy station at time t in scenario i; Let be the load power of the nth load node at time t; For the set of load nodes; Let be the load shedding power of the nth load node at time t for the i-th scenario;

[0046] The following formula is used as the tie-line constraint for the interaction between the microgrid and the main grid:

[0047] In the formula Let be the interaction power of the link line at time t; This is the minimum power allowed for the tie line; This is the maximum power allowed for the tie line; Let be a binary variable representing the microgrid's power purchase status at time t. If the microgrid is in a power purchase status at time t, then... If the microgrid is not in a power purchase state at time t, then ; Let be a binary variable representing the microgrid's electricity sales status at time t. If the microgrid is in the electricity sales status at time t, then... If the microgrid is not in a power sales state at time t, then ;

[0048] Solving the model constructed above yields the interaction power of the link lines at various time periods. And serve as the optimal load baseline.

[0049] Step S4, which involves constructing and solving the second-stage capacity application decision model for the optimal scheduling of the target microgrid based on the results obtained in step S3, specifically includes the following steps:

[0050] After considering microgrid peak shaving, the total cost of the microgrid, including the microgrid peak shaving revenue, is used as the objective function. Based on the constraints of the first-stage load baseline setting model, peak shaving capacity / revenue constraints are added to construct the second-stage application capacity decision model for the optimal scheduling of the target microgrid, and then the model is solved.

[0051] Step S4 specifically includes the following steps:

[0052] Objective function:

[0053] The following formula is used as the objective function of the second-stage capacity application decision model for the optimal scheduling of the target microgrid:

[0054] In the formula This represents the objective function value of the second-stage capacity decision-making model. For the peak-shaving revenue of microgrids, and , For the application of peak-shaving capacity for microgrids, For the expected peak-shaving capacity of the microgrid, , For peak-shaving periods, The optimization objective of the second-stage capacity application decision model is... The optimal load baseline obtained from the model is used to determine the load baseline for the first phase. This is an indicator variable for the peak-shaving type of the microgrid; if the peak-shaving type of the microgrid is peak reduction, then... The peak-shaving type that microgrids participate in is valley filling. , The price is for peak shaving incentives;

[0055] because The calculation formula is a logical expression, therefore the Big-M method is used to... The calculation formula is linearized and expressed as:

[0056] In the formula It is the first largest constant in the big-M model; These are the binary slack variables corresponding to the first logical constraint. The set error tolerance; It is the second largest constant in the big-M class; For the binary slack variables corresponding to the second logical constraint; For microgrid peak shaving revenue; For the binary slack variables corresponding to the third logical constraint;

[0057] Solving the model constructed above yields the declared peak-shaving capacity of the microgrid. This will be used as the final optimized scheduling result.

[0058] This invention also provides a system for implementing the microgrid optimal scheduling method considering grid peak shaving, comprising a data acquisition module, a scenario construction module, a first model construction module, a second model construction module, and an optimal scheduling module; the data acquisition module, scenario construction module, first model construction module, second model construction module, and optimal scheduling module are connected in series; the data acquisition module is used to acquire data information of the target microgrid and the power system, and upload the data information to the scenario construction module; the scenario construction module is used to construct a day-ahead renewable energy power scenario set of the target microgrid based on the received data information and the acquired data information, and upload the data information to the first model construction module; the first model construction module is used to construct and solve a first-stage load baseline formulation model for optimal scheduling of the target microgrid based on the constructed scenario set and the received data information, and upload the data information to the second model construction module; the second model construction module is used to construct and solve a second-stage application capacity decision model for optimal scheduling of the target microgrid based on the received data information and the obtained results, and upload the data information to the optimal scheduling module; the optimal scheduling module is used to complete the optimal scheduling of the target microgrid considering grid peak shaving based on the received data information and the obtained results.

[0059] The microgrid optimization scheduling method and system considering grid peak shaving provided by this invention realizes the randomness of new energy sources by constructing a set of day-ahead new energy power scenarios for the microgrid, and by constructing and solving a two-stage optimization model, it not only realizes the optimization scheduling process of the microgrid considering grid peak shaving, but also has higher reliability and better accuracy. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the method flow of the present invention.

[0061] Figure 2 This is a schematic diagram of the functional modules of the system of the present invention. Detailed Implementation

[0062] like Figure 1 The diagram shown is a flowchart of the method of the present invention: The microgrid optimization scheduling method considering power grid peak shaving disclosed in this invention includes the following steps:

[0063] S1. Acquire data information about the target microgrid and power system;

[0064] S2. Based on the data obtained in step S1, construct the day-ahead renewable energy power scenario set for the target microgrid; including the following steps:

[0065] Obtain the predicted and actual data of each distributed renewable energy source under the target microgrid during each scheduling period;

[0066] Based on the Cornish-Fisher series, a cumulative probability function for the prediction error of renewable energy power in each scheduling period is constructed.

[0067] Based on the sorting method that takes into account time correlation, the prediction error scenarios for each scheduling period are obtained;

[0068] By superimposing the day-ahead renewable energy forecast data for each scheduling period with the forecast error scenario, the day-ahead renewable energy power scenario set for each period of the target microgrid is obtained.

[0069] In practice, the following steps can be taken:

[0070] Obtain the predicted and actual power data of each distributed renewable energy source under the target microgrid during each scheduling period; calculate the power prediction error:

[0071] In the formula Let be the power prediction error of the m-th new energy power station at time t on the j-th historical day; Let be the actual power output of the m-th new energy power station at time t on the j-th historical day; Let m be the predicted power of the m-th new energy power plant at time t on the j-th historical day; , For the collection of new energy power stations; , For time sets; , A collection of historical days;

[0072] Using a Cornish-Fisher series containing fifth-order cumulants, the cumulative probability function of the historical power prediction error for each renewable energy power plant is constructed:

[0073] In the formula Let be the first-order raw moment of the historical prediction error of the m-th new energy power station at time t; The total number of historical days; Let be the second-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the third-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the fourth-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the fifth-order original moment of the historical prediction error of the m-th new energy power station at time t; Let q be the quantile of the probability value q corresponding to the standardized prediction error CDF of the m-th new energy power station at time t. q represents the quantile of the probability value q corresponding to the standard Gaussian distribution; Let q be the quantile of the probability value q corresponding to the standardization prediction error (CDF) of the m-th new energy power station at time t.

[0074] Based on Latin hypercube sampling, the probability q in the cumulative probability function of the historical power prediction error for each new energy power station is set as follows: To obtain the cumulative probability function vector of the prediction error of new energy power in each time period;

[0075] The cumulative probability function vectors of the new energy power prediction errors obtained in each time period are concatenated to obtain the cumulative probability function vector matrix of errors. , represented as:

[0076] Will As the initial set of day-ahead prediction error scenarios;

[0077] Each row of elements monotonically increases from left to right. Clearly, this linearly monotonically structured set of scenarios is insufficient to effectively characterize the random fluctuations of actual errors; to accurately model the temporal correlation characteristics of wind power, it is necessary to reconstruct an error scenario set with equivalent temporal correlation.

[0078] right The permutation and combination of each row of elements in the matrix ensures that the final generated scene set matches the time correlation coefficient of the historical error matrix:

[0079] Calculations were performed on each new energy power station at any time. and Time correlation coefficient between historical error samples for ;

[0080] right The scene vector at each time step undergoes a set of corresponding elementary transformation matrices. Calculate and obtain the power prediction error scenario vector after sorting each time period. for:

[0081] The power prediction error scenario vector for all time periods is calculated using the following formula. The time correlation coefficient of the error scenario matrix:

[0082] In the formula For the m-th new energy power station and The time correlation coefficient of the error scenario matrix between time points; For the m-th new energy power station The power prediction error for the i-th scenario at time i; A collection of scenes; This represents the total number of scenes;

[0083] Logically, it should be implemented right Good tracking; therefore, with As input parameters to the model Assuming the decision variables, construct the optimization problem:

[0084] Solving the constructed optimization problem yields the elementary transformation matrix. This leads to the power prediction error scenario vector sorted by time period. When solving the problem, conventional intelligent algorithms can be used to achieve efficient solutions (differential evolution algorithm, particle swarm optimization algorithm, etc.).

[0085] Finally, the power prediction error scenario vector obtained by superimposing the day-ahead power prediction values ​​of each renewable energy power plant is used to obtain the day-ahead renewable energy power scenario set, denoted as:

[0086] In the formula Let i be the power result of the m-th new energy power station at time t for the i-th scenario; The set that is composed of the current new energy power scenario set; Let be the day-ahead predicted renewable energy power of the m-th renewable energy power station at time t;

[0087] S3. Based on the scenario set constructed in step S2, construct and solve the first-stage load baseline determination model for the optimal scheduling of the target microgrid; specifically including the following steps:

[0088] Without considering microgrid peak shaving, a first-stage load baseline determination model for the optimal scheduling of the target microgrid is constructed and solved, with the total cost of the microgrid as the objective function and the energy storage constraints, power balance constraints, and tie-line constraints of the microgrid as constraints.

[0089] In practice, the following steps can be taken:

[0090] Objective function:

[0091] The following formula is used as the objective function for the first-stage load baseline determination model of the target microgrid's optimal scheduling:

[0092] In the formula Develop the objective function value for the model to determine the first-stage load baseline; The cost of purchasing and selling electricity from the main grid to the microgrid, and , Let be the electricity purchase price of the microgrid at time t. Let be the power purchased by the microgrid at time t. Let be the electricity price of the microgrid at time t. Let be the electricity sold by the microgrid at time t; The cost of energy storage charging and discharging in microgrids, and , A collection of energy storage power stations, Let be the charging and discharging cost coefficient of the e-th energy storage power station. Let e ​​be the charging power of the e-th energy storage power station at time t. Let e ​​be the charging efficiency of the e-th energy storage power station. Let be the discharge power of the e-th energy storage power station at time t. Let be the discharge efficiency of the e-th energy storage power station; The cost of curtailment in microgrids, and , Let i be the probability of the i-th scenario occurring. Let be the curtailment penalty cost coefficient for the m-th renewable energy power station. Let represent the power curtailment of the m-th renewable energy power station at time t in scenario i. Let be the load shedding penalty cost coefficient for the m-th renewable energy power station. Let be the load shedding power of the m-th new energy power station in scenario i at time t.

[0093] Constraints:

[0094] The following formula is used as the energy storage constraint for the microgrid:

[0095] In the formula Let e ​​be the minimum allowable discharge power of the e-th energy storage power station; Let be a binary variable representing the discharge state of the e-th energy storage power station at time t, and let the e-th energy storage power station be in a discharge state at time t. If the e-th energy storage power station is not in a discharging state at time t, then... ; This represents the maximum allowable discharge power of the e-th energy storage power station; The minimum allowed charging power for the e-th energy storage power station; Let be a binary variable representing the charging state of the e-th energy storage station at time t, and let be the charging state of the e-th energy storage station at time t. If the e-th energy storage station is not in a charging state at time t, then... ; This represents the maximum allowed charging power for the e-th energy storage power station. Let be the remaining power of the e-th energy storage power station at time t+1; The minimum allowable storage capacity for the e-th energy storage power station; The maximum allowable storage capacity for the e-th energy storage power station; The remaining power of the e-th energy storage station at the end of the day; The remaining power of the e-th energy storage station at the start of the day; Let be the depth of discharge coefficient of the e-th energy storage power station; Let be the installed capacity of the e-th energy storage power station; where the first constraint gives the boundary conditions for the energy storage charging and discharging power; the second constraint indicates that the energy storage cannot charge and discharge simultaneously in each scheduling period; the third constraint gives the calculation expression for the remaining energy storage capacity, considering that the total number of scheduling periods per day is usually 96 (the scheduling time scale is 15 minutes), "0.25" is introduced to represent the conversion coefficient between power and energy; the fourth constraint is the boundary condition for the remaining energy storage capacity; the fifth constraint stipulates that the remaining energy storage capacity at the beginning and end of the day must be equal to the intermediate value to ensure its reasonable continuous use; given the impact of the depth of discharge on the lifespan of energy storage, the sixth constraint stipulates that the total daily discharge of energy storage must be lower than a certain depth of discharge value;

[0096] The following formula is used as the power balance constraint for the microgrid:

[0097] In the formula Let be the power of the m-th renewable energy station at time t in scenario i; Let be the load power of the nth load node at time t; For the set of load nodes; Let be the load shedding power of the nth load node at time t for the i-th scenario;

[0098] The following formula is used as the tie-line constraint for the interaction between the microgrid and the main grid:

[0099] In the formula Let be the interaction power of the link line at time t; This is the minimum power allowed for the tie line; This is the maximum power allowed for the tie line; Let be a binary variable representing the microgrid's power purchase status at time t. If the microgrid is in a power purchase status at time t, then... If the microgrid is not in a power purchase state at time t, then ; Let be a binary variable representing the microgrid's electricity sales status at time t. If the microgrid is in the electricity sales status at time t, then... If the microgrid is not in a power sales state at time t, then The first constraint specifies the boundary conditions for the interconnected power of the tie line; the second constraint specifies the boundary conditions for the microgrid to purchase / sell electricity; and the third constraint stipulates that the microgrid cannot purchase / sell electricity simultaneously.

[0100] Solving the model constructed above yields the interaction power of the link lines at various time periods. The model is used as the optimal load baseline. When solving the problem, the model is a conventional mixed integer linear programming (MILP) problem, which can be solved efficiently using mature commercial solvers such as GUROBI and CPLEX.

[0101] S4. Based on the results obtained in step S3, construct and solve the second-stage capacity application decision model for the optimal scheduling of the target microgrid; specifically, this includes the following steps:

[0102] After considering microgrid peak shaving, the total cost of the microgrid, including the microgrid peak shaving revenue, is used as the objective function. Based on the constraints of the first-stage load baseline setting model, peak shaving capacity / revenue constraints are added to construct the second-stage application capacity decision model for the optimal scheduling of the target microgrid, and then the model is solved.

[0103] In practice, the following steps can be taken:

[0104] Objective function:

[0105] The following formula is used as the objective function of the second-stage capacity application decision model for the optimal scheduling of the target microgrid:

[0106] In the formula This represents the objective function value of the second-stage capacity decision-making model. For the peak-shaving revenue of microgrids, and , For the application of peak-shaving capacity for microgrids, For the expected peak-shaving capacity of the microgrid, , For peak-shaving periods, The optimization objective of the second-stage capacity application decision model is... The optimal load baseline obtained from the model is used to determine the load baseline for the first phase. This is an indicator variable for the peak-shaving type of the microgrid; if the peak-shaving type of the microgrid is peak reduction, then... The peak-shaving type that microgrids participate in is valley filling. , The price is for peak shaving incentives;

[0107] because The calculation formula is a logical expression, therefore the Big-M method is used to... The calculation formula is linearized and expressed as:

[0108] In the formula It is the first largest constant in the big-M model; These are the binary slack variables corresponding to the first logical constraint. The set error tolerance; It is the second largest constant in the big-M class; For the binary slack variables corresponding to the second logical constraint; For microgrid peak shaving revenue; For the binary slack variables corresponding to the third logical constraint;

[0109] Solving the model constructed above yields the declared peak-shaving capacity of the microgrid. The result is used as the final optimized scheduling result. The model here is a holistic model that includes the first and second stages. This model is also a MILP problem and can be solved efficiently using mature commercial solvers such as GUROBI and CPLEX.

[0110] S5. Based on the results obtained in steps S3 and S4, complete the optimized scheduling of the target microgrid considering grid peak shaving.

[0111] This invention designs a two-stage stochastic dispatch strategy for microgrids that takes into account grid peak shaving. Compared to traditional methods (such as microgrid dispatch strategies that do not consider grid peak shaving, and methods that obtain load baselines based on weighted historical data), the microgrid can flexibly optimize the declared load baseline and peak shaving capacity based on the latest renewable energy power forecast results during the day-ahead phase, effectively balancing economy, safety, and reliability. The error CDF fitting method based on Cornish-Fisher series in this invention can directly realize the explicit functional mapping from samples to CDF, skipping the probability distribution acquisition step compared to parametric methods such as Gaussian distribution and general distribution, resulting in more efficient computation. The scenario reordering method that takes into account time correlation can fully exploit the inherent temporal coupling characteristics of error data, making the generated renewable energy power scenarios more consistent with the temporal characteristics of historical samples. Therefore, the method of this invention has higher reliability and better accuracy.

[0112] like Figure 2The diagram shows the functional modules of the system of the present invention: The system disclosed in this invention for implementing the microgrid optimal scheduling method considering grid peak shaving includes a data acquisition module, a scenario construction module, a first model construction module, a second model construction module, and an optimal scheduling module; the data acquisition module, scenario construction module, first model construction module, second model construction module, and optimal scheduling module are connected in series; the data acquisition module is used to acquire data information of the target microgrid and the power system, and upload the data information to the scenario construction module; the scenario construction module is used to construct a day-ahead renewable energy power scenario set of the target microgrid based on the received data information and the acquired data information, and upload the data information to the first model construction module; the first model construction module is used to construct and solve a first-stage load baseline formulation model for optimal scheduling of the target microgrid based on the constructed scenario set and the received data information, and upload the data information to the second model construction module; the second model construction module is used to construct and solve a second-stage application capacity decision model for optimal scheduling of the target microgrid based on the received data information and the obtained results, and upload the data information to the optimal scheduling module; the optimal scheduling module is used to complete the optimal scheduling of the target microgrid considering grid peak shaving based on the received data information and the obtained results.

Claims

1. A microgrid optimal scheduling method considering power grid peak shaving, comprising the following steps: S1. Acquire data information about the target microgrid and power system; S2. Based on the data obtained in step S1, construct a set of day-ahead renewable energy power scenarios for the target microgrid; S3. Based on the scenario set constructed in step S2, construct and solve the first-stage load baseline formulation model for the optimal scheduling of the target microgrid; S4. Based on the results obtained in step S3, construct and solve the second-stage capacity application decision model for the optimal scheduling of the target microgrid; S5. Based on the results obtained in steps S3 and S4, complete the optimized scheduling of the target microgrid considering grid peak shaving.

2. The microgrid optimal scheduling method considering grid peak shaving according to claim 1, characterized in that... Step S2, which involves constructing a day-ahead renewable energy power scenario set for the target microgrid based on the data information obtained in step S1, includes the following steps: Obtain the predicted and actual data of each distributed renewable energy source under the target microgrid during each scheduling period; Based on the Cornish-Fisher series, a cumulative probability function for the prediction error of renewable energy power in each scheduling period is constructed. Based on the sorting method that takes into account time correlation, the prediction error scenarios for each scheduling period are obtained; By overlaying the day-ahead renewable energy forecast data for each scheduling period with the forecast error scenario, the day-ahead renewable energy power scenario set for each period of the target microgrid is obtained.

3. The microgrid optimal scheduling method considering grid peak shaving according to claim 2, characterized in that... Step S2 specifically includes the following steps: Obtain the predicted and actual power data of each distributed renewable energy source under the target microgrid during each scheduling period; calculate the power prediction error: In the formula Let be the power prediction error of the m-th new energy power station at time t on the j-th historical day; Let be the actual power output of the m-th new energy power station at time t on the j-th historical day; Let m be the predicted power of the m-th new energy power plant at time t on the j-th historical day; , For the collection of new energy power stations; , For time sets; , A collection of historical days; Using a Cornish-Fisher series containing fifth-order cumulants, the cumulative probability function of the historical power prediction error for each renewable energy power plant is constructed: In the formula Let be the first-order raw moment of the historical prediction error of the m-th new energy power station at time t; The total number of historical days; Let be the second-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the third-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the fourth-order original moment of the historical prediction error of the m-th new energy power station at time t; Let be the fifth-order original moment of the historical prediction error of the m-th new energy power station at time t; Let q be the quantile of the probability value q corresponding to the standardized prediction error CDF of the m-th new energy power station at time t. q represents the quantile of the probability value q corresponding to the standard Gaussian distribution; Let q be the quantile of the probability value q corresponding to the standardization prediction error (CDF) of the m-th new energy power station at time t. Based on Latin hypercube sampling, the probability q in the cumulative probability function of the historical power prediction error for each new energy power station is set as follows: To obtain the cumulative probability function vector of the prediction error of new energy power in each time period; The cumulative probability function vectors of the new energy power prediction errors obtained in each time period are concatenated to obtain the cumulative probability function vector matrix of errors. , represented as: Will As the initial set of day-ahead prediction error scenarios; right The permutation and combination of each row of elements in the matrix ensures that the final generated scene set matches the time correlation coefficient of the historical error matrix: Calculations were performed on each new energy power station at any time. and Time correlation coefficient between historical error samples for ; right The scene vector at each time step undergoes a set of corresponding elementary transformation matrices. Calculate and obtain the power prediction error scenario vector after sorting each time period. for: The power prediction error scenario vector for all time periods is calculated using the following formula. The time correlation coefficient of the error scenario matrix: In the formula For the m-th new energy power station and The time correlation coefficient of the error scenario matrix between time points; For the m-th new energy power station The power prediction error for the i-th scenario at time i; A collection of scenes; This represents the total number of scenes; Optimization problem: Solving the constructed optimization problem yields the elementary transformation matrix. This leads to the power prediction error scenario vector sorted by time period. ; Finally, the power prediction error scenario vector obtained by superimposing the day-ahead power prediction values ​​of each renewable energy power plant is used to obtain the day-ahead renewable energy power scenario set, denoted as: In the formula Let i be the power result of the m-th new energy power station at time t for the i-th scenario; The set that is composed of the current new energy power scenario set; Let t be the day-ahead predicted power of the m-th renewable energy power station at time t.

4. The microgrid optimal scheduling method considering grid peak shaving according to claim 3, characterized in that... Step S3, based on the scenario set constructed in step S2, involves constructing and solving the first-stage load baseline determination model for the optimal scheduling of the target microgrid. Specifically, this includes the following steps: Without considering microgrid peak shaving, a first-stage load baseline determination model for the optimal scheduling of the target microgrid is constructed and solved, with the total cost of the microgrid as the objective function and the energy storage constraints, power balance constraints, and tie-line constraints of the microgrid as constraints.

5. The microgrid optimal scheduling method considering grid peak shaving according to claim 4, characterized in that... Step S3 specifically includes the following steps: Objective function: The following formula is used as the objective function for the first-stage load baseline determination model of the target microgrid's optimal scheduling: In the formula Develop the objective function value for the model to determine the first-stage load baseline; The cost of purchasing and selling electricity from the main grid to the microgrid, and , Let be the electricity purchase price of the microgrid at time t. Let be the power purchased by the microgrid at time t. Let be the electricity price of the microgrid at time t. Let be the electricity sold by the microgrid at time t; The cost of energy storage charging and discharging in microgrids, and , A collection of energy storage power stations, Let be the charging and discharging cost coefficient of the e-th energy storage power station. Let e ​​be the charging power of the e-th energy storage power station at time t. Let e ​​be the charging efficiency of the e-th energy storage power station. Let be the discharge power of the e-th energy storage power station at time t. Let be the discharge efficiency of the e-th energy storage power station; The cost of curtailment in microgrids, and , Let i be the probability of the i-th scenario occurring. Let be the curtailment penalty cost coefficient for the m-th renewable energy power station. Let represent the power curtailment of the m-th renewable energy power station at time t in scenario i. Let be the load shedding penalty cost coefficient for the m-th renewable energy power station. Let be the load shedding power of the m-th new energy power station in scenario i at time t. Constraints: The following formula is used as the energy storage constraint for the microgrid: In the formula Let e ​​be the minimum allowable discharge power of the e-th energy storage power station; Let be a binary variable representing the discharge state of the e-th energy storage power station at time t, and let the e-th energy storage power station be in a discharge state at time t. If the e-th energy storage power station is not in a discharging state at time t, then... ; This represents the maximum allowable discharge power of the e-th energy storage power station; The minimum allowed charging power for the e-th energy storage power station; Let be a binary variable representing the charging state of the e-th energy storage station at time t, and let be the charging state of the e-th energy storage station at time t. If the e-th energy storage station is not in a charging state at time t, then... ; This represents the maximum allowed charging power for the e-th energy storage power station. Let be the remaining power of the e-th energy storage power station at time t+1; The minimum allowable storage capacity for the e-th energy storage power station; The maximum allowable storage capacity for the e-th energy storage power station; The remaining power of the e-th energy storage station at the end of the day; The remaining power of the e-th energy storage station at the start of the day; Let be the depth of discharge coefficient of the e-th energy storage power station; Let be the installed capacity of the e-th energy storage power station; The following formula is used as the power balance constraint for the microgrid: In the formula Let be the power of the m-th renewable energy station at time t in scenario i; Let be the load power of the nth load node at time t; For the set of load nodes; Let be the load shedding power of the nth load node at time t for the i-th scenario; The following formula is used as the tie-line constraint for the interaction between the microgrid and the main grid: In the formula Let be the interaction power of the link line at time t; This is the minimum power allowed for the tie line; This is the maximum power allowed for the tie line; Let be a binary variable representing the microgrid's power purchase status at time t. If the microgrid is in a power purchase status at time t, then... If the microgrid is not in a power purchase state at time t, then ; Let be a binary variable representing the microgrid's electricity sales status at time t. If the microgrid is in the electricity sales status at time t, then... If the microgrid is not in a power sales state at time t, then ; Solving the model constructed above yields the interaction power of the link lines at various time periods. And serve as the optimal load baseline.

6. The microgrid optimal scheduling method considering grid peak shaving according to claim 5, characterized in that... Step S4, which involves constructing and solving the second-stage capacity application decision model for the optimal scheduling of the target microgrid based on the results obtained in step S3, specifically includes the following steps: After considering microgrid peak shaving, the total cost of the microgrid, including the microgrid peak shaving revenue, is used as the objective function. Based on the constraints of the first-stage load baseline setting model, peak shaving capacity / revenue constraints are added to construct the second-stage application capacity decision model for the optimal scheduling of the target microgrid, and then the model is solved.

7. The microgrid optimal scheduling method considering grid peak shaving according to claim 6, characterized in that... Step S4 specifically includes the following steps: Objective function: The following formula is used as the objective function of the second-stage capacity application decision model for the optimal scheduling of the target microgrid: In the formula This represents the objective function value of the second-stage capacity decision-making model. For the peak-shaving revenue of microgrids, and , For the application of peak-shaving capacity for microgrids, For the expected peak-shaving capacity of the microgrid, , For peak-shaving periods, The optimization objective of the second-stage capacity application decision model is... The optimal load baseline obtained from the model is used to determine the load baseline for the first phase. This is an indicator variable for the peak-shaving type of the microgrid; if the peak-shaving type of the microgrid is peak reduction, then... The peak-shaving type that microgrids participate in is valley filling. , The price is for peak shaving incentives; because The calculation formula is a logical expression, therefore the Big-M method is used to... The calculation formula is linearized and expressed as: In the formula It is the first largest constant in the big-M model; These are the binary slack variables corresponding to the first logical constraint. The set error tolerance; It is the second largest constant in the big-M class; For the binary slack variables corresponding to the second logical constraint; For microgrid peak shaving revenue; For the binary slack variables corresponding to the third logical constraint; Solving the model constructed above yields the declared peak-shaving capacity of the microgrid. This will be used as the final optimized scheduling result.

8. A system for implementing the microgrid optimal scheduling method considering grid peak shaving as described in any one of claims 1 to 7, characterized in that... It includes a data acquisition module, a scenario construction module, a first model construction module, a second model construction module, and an optimization scheduling module; the data acquisition module, scenario construction module, first model construction module, second model construction module, and optimization scheduling module are connected in series; the data acquisition module is used to acquire data information of the target microgrid and power system, and upload the data information to the scenario construction module; the scenario construction module is used to construct the day-ahead renewable energy power scenario set of the target microgrid based on the received data information and the acquired data information, and upload the data information to the first model construction module; The first model building module is used to construct the first-stage load baseline of the target microgrid's optimized scheduling based on the received data information and the constructed scenario set, formulate the model and solve it, and upload the data information to the second model building module. The second model building module is used to construct and solve the second-stage application capacity decision model for the optimized scheduling of the target microgrid based on the received data information and the obtained results, and upload the data information to the optimized scheduling module. The optimization scheduling module is used to perform optimized scheduling of the target microgrid, taking into account grid peak shaving, based on the received data and the obtained results.

Images