A dispatching method for offshore wind power energy storage participating in peak shaving auxiliary service

Abstract

The application discloses a kind of offshore wind power energy storage participates in peak shaving auxiliary service scheduling method, the method provided by the present application, by generating dispatching scheme initial solution, and based on initial solution iteration variation to seek the solution that can make the objective function value optimal to meet constraint condition, dispatching scheme includes the wind turbine generator power of each period in target date, energy storage system charge-discharge state, the charge-discharge power of energy storage system and power consumption data, so that the power consumption data of power grid can be reasonably peak shaving, and the excess power can be stored using energy storage system charging when wind power is surplus, and energy storage system is discharged to realize supplement when wind power is insufficient, improve the utilization of wind power resources.

Description

Technical Field

[0001] This invention relates to the field of new energy technology, and in particular to a scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services. Background Technology

[0002] Offshore wind resources exhibit uneven distribution between day and night, with wind resources being more abundant at night. However, current technologies do not fully utilize the uneven distribution of wind resources during the day for peak shaving assistance, resulting in low utilization of wind power resources and a significant waste of wind power resources.

[0003] Therefore, existing technologies still need to be improved and enhanced. Summary of the Invention

[0004] To address the aforementioned deficiencies in existing technologies, this invention provides a scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services, aiming to solve the problem of low utilization rate of wind power generation resources in existing technologies.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A first aspect of the present invention provides a scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services, the method comprising:

[0007] Obtain initialization information, which includes wind power forecast information and electricity load demand information for the target date. Based on the initialization information, randomly generate initial solutions for multiple scheduling schemes. Add the initial solutions to the target set. Perform constraint processing on the solutions in the target set based on pre-set power grid constraints to obtain the updated target set.

[0008] For each solution in the target set, two solutions are randomly selected from its neighborhood, and a first solution is generated based on the two solutions.

[0009] The first solution is mutated multiple times based on the optimal sub-objective function value among the solutions in the current target set to generate a second solution. The target set is then updated based on the objective function value corresponding to the second solution and the objective function value corresponding to the solution in the current target set.

[0010] The process of re-exercising the step of randomly selecting two solutions in the neighborhood of each solution in the target set and performing gene recombination on the two solutions using an adaptive crossover operator is repeated until a preset number of iterations is reached.

[0011] After reaching the preset number of iterations, the scheduling scheme corresponding to the optimal solution is selected from the target set as the target scheduling scheme for the target date. The wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power and electricity consumption data are controlled according to the target scheduling scheme for each time period of the target date.

[0012] Each of the scheduling schemes includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date.

[0013] The objective function value includes multiple sub-objective function values, which include a first sub-objective function value and a second sub-objective function value. The first sub-objective function value is obtained based on the energy market revenue, and the second sub-objective function value is obtained based on a preset wind curtailment penalty calculation formula.

[0014] The scheduling method for offshore wind power energy storage to participate in peak shaving auxiliary services, wherein the sub-objective function value further includes: a third sub-objective function value and a fourth sub-objective function value;

[0015] The value of the third sub-objective function is obtained based on the preset loss cost of the energy storage system;

[0016] The value of the fourth sub-objective function is obtained based on the variance of the output power fluctuation of the wind turbine and the energy storage system.

[0017] The aforementioned scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services, wherein obtaining initialization information includes:

[0018] Obtain wind forecast data for the target date;

[0019] The wind forecast data is input into a pre-trained neural network model to obtain the wind power prediction information output by the neural network model.

[0020] The aforementioned method for dispatching offshore wind power storage to participate in peak-shaving ancillary services, wherein the grid constraints include:

[0021] The wind power generation capacity does not exceed the wind power prediction information;

[0022] The charging and discharging power of the energy storage system shall not exceed the upper limit of the charging and discharging power of the energy storage system.

[0023] The fluctuations in wind power generation and the charging and discharging power of energy storage systems during different time periods shall not exceed the upper limit of fluctuation.

[0024] User consumption data within each time period does not exceed the pre-set power consumption limits for each user.

[0025] User consumption data within each time period satisfies the user load demand information.

[0026] The aforementioned scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services, wherein generating a first solution based on the two solutions includes:

[0027] An adaptive crossover operator is used to recombine the two solutions to generate the first solution.

[0028] The calculation formula for the adaptive crossover operator is:

[0029] PC i =a min +a max (i / N);

[0030] Where pc is the adaptive crossover operator in the i-th iteration, and a min For the preset minimum crossover probability, a max Where N is the preset maximum crossover probability, and N is the preset number of iterations.

[0031] The aforementioned scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services, wherein the step of performing multiple mutations on the first solution based on the optimal sub-objective function value among the solutions of the current objective set to generate a second solution includes:

[0032] Based on the sub-objective function values ​​corresponding to all solutions in the current objective set, determine the optimal sub-objective function values;

[0033] Based on the optimal values ​​of each sub-objective function, the optimal distance corresponding to the first solution is determined by the first preset formula;

[0034] Based on the optimal distance corresponding to the first solution, the second solution is generated using a second preset formula during each mutation.

[0035] The first preset formula is:

[0036]

[0037] The optimal distance corresponds to the mutation result of the i-th first solution in the t-th iteration. f is the nth sub-objective function value corresponding to the tth mutation result of the i-th first solution. nmin This represents the optimal value of the nth sub-objective function in the current set of objectives.

[0038] The second preset formula is:

[0039]

[0040] stepsize is a variable asynchronous length. Let represent the result of the t-th mutation of the i-th first solution. Let represent the mutation result of the i-th first solution in the (t+1)-th iteration, where randon is a random number that follows a standard normal distribution, s(λ) is the random search path, and α is a scaling factor. This indicates a dot product operation. When the number of mutations reaches the preset number of mutations, the mutation stops, and the final mutation result is used as the second solution.

[0041] The aforementioned scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services, wherein updating the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the objective solution in the current target set includes:

[0042] For each of the second solutions, the following processing is performed:

[0043] The third solution is removed from the objective set according to the second solution, wherein each sub-objective function value of the third solution is worse than the corresponding sub-objective function value of the second solution;

[0044] If the fourth solution does not exist in the target set, the second solution is added to the target set, and each sub-objective function value in the fourth section is better than the sub-objective function value of the corresponding second solution.

[0045] A second aspect of the present invention provides a scheduling device for offshore wind power energy storage to participate in peak shaving ancillary services, comprising:

[0046] An initialization module is used to acquire initialization information, including wind power forecast information and electricity load demand information for a target date. Based on the initialization information, multiple initial solutions for scheduling schemes are randomly generated, and the initial solutions are added to a target set. Based on pre-set constraints, the individuals in the target set are constrained to obtain an updated target set.

[0047] The first mutation module is used to randomly select two solutions in the neighborhood of each target solution in the target set, and generate a first solution based on the two solutions.

[0048] The second mutation module is used to perform multiple mutations on the first solution based on the optimal sub-objective function value among the solutions of the current target set to generate a second solution, and update the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the target solution in the current target set.

[0049] An iterative module is used to re-execute the step of randomly selecting two solutions in the neighborhood of each target solution in the target set and performing gene recombination on the two solutions using an adaptive crossover operator until a preset number of iterations is reached.

[0050] The scheduling module is used to select the scheduling scheme corresponding to the optimal solution from the target set as the target scheduling scheme for the target date after reaching a preset number of iterations, and to control the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power and electricity consumption data in each time period of the target date according to the target scheduling scheme.

[0051] Each of the scheduling schemes includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date.

[0052] The objective function value includes multiple sub-objective function values, which include a first sub-objective function value and a second sub-objective function value. The first sub-objective function value is obtained based on the energy market revenue, and the second sub-objective function value is obtained based on a preset wind curtailment penalty calculation formula.

[0053] A third aspect of the present invention provides a terminal, the terminal including a processor and a computer-readable storage medium communicatively connected to the processor, the computer-readable storage medium being adapted to store a plurality of instructions, the processor being adapted to invoke the instructions in the computer-readable storage medium to execute the steps of the scheduling method for implementing the offshore wind power energy storage participating in peak shaving ancillary services as described in any of the preceding claims.

[0054] In a fourth aspect, the present invention provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the steps of the scheduling method for offshore wind power storage participating in peak shaving ancillary services as described in any of the preceding claims.

[0055] Compared with existing technologies, this invention provides a scheduling method and related equipment for offshore wind power energy storage to participate in peak shaving ancillary services. The scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services generates an initial solution for the scheduling scheme and performs iterative mutation based on the initial solution to seek a solution that satisfies the constraints and optimizes the objective function value. The scheduling scheme includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date. This allows for reasonable peak shaving of the power grid's electricity consumption data, and can store excess electricity by charging the energy storage system when wind power is surplus, and supplement it by discharging the energy storage system when wind power is insufficient, thereby improving the utilization rate of wind power resources. Attached Figure Description

[0056] Figure 1 A flowchart illustrating an embodiment of the scheduling method for offshore wind power energy storage participating in peak shaving ancillary services provided by the present invention;

[0057] Figure 2 A schematic diagram of the structural principle of an embodiment of the dispatching device for offshore wind power energy storage participating in peak shaving auxiliary services provided by the present invention;

[0058] Figure 3 A schematic diagram illustrating the principle of an embodiment of the terminal provided by the present invention. Detailed Implementation

[0059] To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0060] The scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services provided by the present invention can be applied to terminals with computing capabilities. The terminals can execute the scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services provided by the present invention to perform grid peak shaving scheduling. The terminals can be, but are not limited to, various computers, mobile terminals, smart home appliances, wearable devices, etc.

[0061] Example 1

[0062] like Figure 1 As shown, one embodiment of the scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services includes the following steps:

[0063] S100. Obtain initialization information, which includes wind power forecast information and electricity load demand information for the target date. Based on the initialization information, generate initial solutions for multiple scheduling schemes randomly, add the initial solutions to the target set, and perform constraint processing on individuals in the target set based on pre-set power grid constraints to obtain the updated target set.

[0064] Specifically, the scheduling scheme includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date. The method provided in this embodiment targets a wind power grid, which includes wind turbine generators for power generation. Furthermore, to improve the utilization rate of wind power resources, an energy storage system is also provided, which may include supercapacitors and batteries. The purpose of the method provided in this embodiment is to find a scheduling scheme that schedules the wind turbine power generation, energy storage system charging and discharging status, charging and discharging power, and electricity consumption data according to this scheme, so that the grid can meet the electricity load demand of users and achieve the optimal objective function value as much as possible.

[0065] In one possible implementation, considering better economic efficiency, the objective function value includes multiple sub-objective function values, namely: a first sub-objective function value and a second sub-objective function value. The first sub-objective function value is obtained based on energy market revenue, and the second sub-objective function value is obtained based on a preset wind curtailment penalty calculation formula. Specifically, based on economic efficiency considerations, during scheduling, it is desirable to use more wind power to directly meet users' electricity load, i.e., directly supply power to users, rather than prioritizing electricity storage. Therefore, the first sub-objective function value is obtained based on energy market revenue. On the other hand, it is also desirable to utilize all wind power as much as possible and avoid wasting wind resources. Therefore, a wind curtailment penalty calculation formula is preset. The more wind resources are abandoned, the greater the wind curtailment penalty. To unify the optimization direction of the various sub-objective function values, in this embodiment, the smaller the objective function value, the better the corresponding solution. Therefore, in this embodiment, the first sub-objective function value is obtained by taking a negative weight on the energy market revenue value, and the second sub-objective function value is obtained by taking a positive weight on the wind curtailment penalty calculation result.

[0066] Specifically, in the method provided in this embodiment, wind power forecast information and electricity load demand information within the target date are first obtained, and multiple initial solutions are generated. Then, the initial solutions are iteratively mutated multiple times to find a better solution. The acquisition of initialization information includes:

[0067] Obtain wind forecast data for the target date;

[0068] The wind forecast data is input into a pre-trained neural network model to obtain the wind power prediction information output by the neural network model.

[0069] In this embodiment, a backpropagation (BP) neural network model is used. Of course, those skilled in the art can choose other neural network model structures. The neural network model is trained using multiple sets of training data, each set of training data including sample wind forecast data and corresponding wind power labeling information.

[0070] The wind forecast data for the target date includes wind forecast information for each time period within the target date, including wind direction and wind speed.

[0071] It is easy to see that in order to optimize the first and second sub-objective function values, it is necessary to prioritize direct power supply to users during nighttime hours when wind resources are more abundant. However, excessive direct power supply to users at night may result in the wind power generated during the day not being consumed by user loads. Furthermore, when the input power of wind turbines is high at night, if wind power fluctuates, the energy storage system needs to discharge rapidly to compensate for the impact of wind power fluctuations, resulting in significant losses to the energy storage system. In addition, some user electricity demands cannot be scheduled at night. Therefore, in this embodiment, to achieve comprehensive optimization of the scheduling scheme, in addition to considering energy market revenue and wind curtailment penalties, the sub-objective function values ​​also include: a third sub-objective function value and a fourth sub-objective function value.

[0072] The value of the third sub-objective function is obtained based on the preset loss cost of the energy storage system;

[0073] The value of the fourth sub-objective function is obtained based on the variance of the output power fluctuation of the wind turbine and the energy storage system.

[0074] The loss cost of the energy storage system should be as low as possible, and the output power of the wind turbine and the energy storage system should also be as stable as possible. Therefore, positive weight values ​​are taken for both the third and fourth sub-objective function values.

[0075] It should be noted that, in this embodiment, the sign of the weight of the sub-objective function value is determined based on the optimization direction of the total objective function value being minimized. When the optimization direction of the total objective function value is maximized, the signs of the weights of each sub-objective function value should be opposite.

[0076] In practical applications, dispatching schemes are limited by the equipment performance and parameters of wind turbines and energy storage systems, as well as the specific electricity demands of users. In other words, dispatching schemes that exceed the equipment performance and parameters of wind turbines and energy storage systems, or that cannot meet specific user electricity demands, are practically impossible to implement. Therefore, after generating the initial solution based on the initial information, the initial solution needs to be constrained. Constraint processing can involve deleting solutions in the target set that do not satisfy the grid constraints, or adjusting them to satisfy the grid constraints. Specifically, the grid constraints include:

[0077] The wind power generation capacity does not exceed the wind power prediction information;

[0078] The charging and discharging power of the energy storage system shall not exceed the upper limit of the charging and discharging power of the energy storage system.

[0079] The fluctuations in wind power generation and the charging and discharging power of energy storage systems during different time periods shall not exceed the upper limit of fluctuation.

[0080] User consumption data within each time period does not exceed the pre-set power consumption limits for each user.

[0081] User consumption data within each time period satisfies the user load demand information.

[0082] The information that user consumption data in each time period meets the user load demand information and that user consumption data in each time period does not exceed the preset power consumption limit information for each user means that user consumption data in each time period should reach the user's total load demand, and the amount of power supplied to the user in each time period should meet the user's power consumption time limit. For example, although wind power resources are abundant at night, not all user consumption data can be scheduled for nighttime hours, because some of the user's power load must be operated during the day.

[0083] Only solutions that satisfy the above power grid constraints can be truly realized in practical applications.

[0084] Please refer to it again. Figure 1 The method provided in this embodiment further includes the following steps:

[0085] S200. For each solution in the target set, randomly select two solutions in its neighborhood, and generate a first solution based on the two solutions.

[0086] S300. Perform multiple mutations on the first solution based on the optimal sub-objective function value among the solutions in the current target set to generate a second solution. Update the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the solution in the current target set.

[0087] S400. Re-execute the step of randomly selecting two solutions in the neighborhood of each target solution in the target set and using an adaptive crossover operator to perform gene recombination on the two solutions until the preset number of iterations is reached.

[0088] In this embodiment, an iterative approach is used to mutate and optimize the solutions in the target set. The execution process in each iteration is as follows:

[0089] First, for each solution in the target set, two solutions are randomly selected from its neighborhood. Specifically, each solution in the target set can be considered as a point in space. If the distance vector between two points is within a preset neighborhood vector range, then these two points are in each other's neighborhood. For each solution in the target set, two solutions are randomly selected from its neighborhood according to the preset neighborhood vector, and a first solution is generated based on these two solutions. In other words, for each solution in the target set, a corresponding first solution is generated.

[0090] In this embodiment, the first solution is generated using gene recombination. Generating a first solution based on the two solutions includes:

[0091] An adaptive crossover operator is used to recombine the two solutions to generate the first solution.

[0092] The calculation formula for the adaptive crossover operator is:

[0093] PC i =a min +a max (i / N);

[0094] Where pc is the adaptive crossover operator in the i-th iteration, and a min For the preset minimum crossover probability, a max Where N is the preset maximum crossover probability, and N is the preset number of iterations.

[0095] It is easy to see that as the number of iterations increases, the adaptive crossover operator becomes larger, which means that the crossover probability of the two solutions will increase, thus expanding the search range for the optimal solution and preventing the omission of better solutions.

[0096] After generating the first solution, the first solution is subjected to multiple mutation processes to generate a second solution. Specifically, the process of performing multiple mutation processes on the first solution based on the optimal sub-objective function value among the solutions in the current objective set to generate the second solution includes:

[0097] Based on the sub-objective function values ​​corresponding to all solutions in the current objective set, determine the optimal sub-objective function values;

[0098] Based on the optimal values ​​of each sub-objective function, the optimal distance corresponding to the first solution is determined by the first preset formula;

[0099] Based on the optimal distance corresponding to the first solution, the second solution is generated using a second preset formula during each mutation.

[0100] The first preset formula is:

[0101]

[0102] The optimal distance corresponds to the mutation result of the i-th first solution in the t-th iteration. f is the nth sub-objective function value corresponding to the tth mutation result of the i-th first solution. nmin This represents the optimal value of the nth sub-objective function in the current set of objectives.

[0103] The second preset formula is:

[0104]

[0105] stepsize is a variable asynchronous length. Let represent the result of the t-th mutation of the i-th first solution. Let represent the mutation result of the i-th first solution in the (t+1)-th iteration, where randon is a random number that follows a standard normal distribution, s(λ) is the random search path, and α is a scaling factor, which can be 1. This indicates a dot product operation. When the number of mutations reaches the preset number of mutations, the mutation stops, and the final mutation result is used as the second solution.

[0106] To prevent premature convergence due to a lack of diversity in the later stages of iteration, which could lead to missing the optimal solution, this embodiment performs multiple mutations on the first solution obtained after gene recombination to search for more solutions. Furthermore, during each mutation, in addition to searching for more solutions through random search paths, to make the mutation direction closer to the direction of the optimal solution, the distance between the first solution and the current ideal optimal solution (the optimal distance) is obtained based on the difference between the sub-objective function values ​​corresponding to the first solution and the optimal sub-objective function values ​​in the current objective set. This optimal distance is then used to mutate the first solution. Each solution can be projected into a vector space. For example, the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data can be used as elements of a vector. Alternatively, an embedding matrix can be used to project a solution into a vector to achieve calculation, i.e., x. it is the vector form corresponding to the i-th mutation result of the first solution in the t-th iteration.

[0107] After reaching the preset number of mutations and obtaining the second solution, instead of directly adding the second solution to the target set, solutions that are significantly inferior to the second solution are eliminated from the target set based on the second solution. Furthermore, to ensure the practical applicability of the generated solution, before updating the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the target solution in the current target set, the following step is also included:

[0108] The second solution is constrained according to the power grid constraints.

[0109] Specifically, updating the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the objective solution in the current target set includes:

[0110] For each of the second solutions, the following processing is performed:

[0111] The third solution is removed from the objective set according to the second solution, wherein each sub-objective function value of the third solution is worse than the corresponding sub-objective function value of the second solution;

[0112] If the fourth solution does not exist in the target set, the second solution is added to the target set, and each sub-objective function value in the fourth section is better than the sub-objective function value of the corresponding second solution.

[0113] For each second solution, they are sorted according to the merits of their corresponding objective function values ​​and processed sequentially. Each time a second solution is processed, the sub-objective function values ​​corresponding to all solutions in the objective set are first determined. If there exists a solution whose sub-objective function values ​​are all worse than the sub-objective function values ​​of the corresponding second solution, then this solution is designated as the third solution. That is, the first sub-objective function value of the third solution is worse than the first sub-objective function value of the second solution, the second sub-objective function value of the third solution is worse than the second sub-objective function value of the second solution, and so on. All third solutions are then removed from the objective set.

[0114] Next, it is determined whether the fourth solution exists in the target set. Specifically, if a solution in the target set has all its sub-objective function values ​​better than the corresponding sub-objective function values ​​of the second solution, then this solution is designated as the fourth solution. That is, the first sub-objective function value of the fourth solution is better than the first sub-objective function value of the second solution, the second sub-objective function value of the fourth solution is better than the second sub-objective function value of the second solution, and so on. If the fourth solution exists, the second solution is not added to the target set; if the fourth solution does not exist, the second solution is added to the target set. This effectively eliminates inferior solutions and retains superior solutions for mutation in the next iteration.

[0115] After processing all the second solutions, this round of iteration ends, and a new round of iteration begins. For each solution in the current target set, the corresponding first solution is regenerated. The iteration ends when the preset number of iterations is reached.

[0116] Please refer to it again. Figure 1 The method provided in this embodiment further includes the following steps:

[0117] S500. After reaching the preset number of iterations, select the scheduling scheme corresponding to the optimal solution from the target set as the target scheduling scheme for the target date, and control the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power and electricity consumption data for each time period of the target date according to the target scheduling scheme.

[0118] After the iteration is completed, the optimal solution is selected from the target set, that is, the scheduling scheme corresponding to the solution with the best objective function value is selected as the target scheduling scheme for the target date, and scheduling is carried out according to the target scheduling scheme within the target date.

[0119] In summary, this embodiment provides a scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services. By generating an initial solution for the scheduling scheme and iteratively mutating it based on the initial solution, a solution that satisfies the constraints and optimizes the objective function value is sought. The scheduling scheme includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date. This allows for reasonable peak shaving of the power grid's electricity consumption data. When there is a surplus of wind power, the excess electricity can be stored by charging the energy storage system, and when there is a shortage of wind power, the excess electricity can be supplemented by discharging the energy storage system, thereby improving the utilization rate of wind power resources.

[0120] It should be understood that although the steps in the flowcharts shown in the accompanying drawings are displayed sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0121] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0122] Example 2

[0123] Based on the above embodiments, the present invention also provides a scheduling device for offshore wind power energy storage to participate in peak shaving ancillary services, such as... Figure 2 As shown, the dispatching device for offshore wind power energy storage participating in peak shaving ancillary services includes:

[0124] An initialization module is used to acquire initialization information, which includes wind power forecast information and electricity load demand information for a target date. Based on the initialization information, multiple initial solutions for scheduling schemes are randomly generated, and the initial solutions are added to the target set. Based on pre-set constraints, the individuals in the target set are constrained to obtain the updated target set, as described in Embodiment 1.

[0125] The first mutation module is used to randomly select two solutions in the neighborhood of each target solution in the target set, and generate a first solution based on the two solutions, as described in Embodiment 1.

[0126] The second mutation module is used to perform multiple mutations on the first solution based on the optimal sub-objective function value among the solutions of the current target set to generate a second solution, and update the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the target solution in the current target set, as specifically described in Embodiment 1;

[0127] An iterative module is used to re-execute the step of randomly selecting two solutions in the neighborhood of each target solution in the target set and performing gene recombination on the two solutions using an adaptive crossover operator until a preset number of iterations is reached, as described in Example 1.

[0128] The scheduling module is used to select the optimal solution corresponding to the scheduling scheme from the target set as the target scheduling scheme for the target date after reaching a preset number of iterations, and to control the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power and electricity consumption data in each time period of the target date according to the target scheduling scheme, as specifically described in Embodiment 1;

[0129] Each of the scheduling schemes includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date, as described in Example 1.

[0130] The objective function value includes multiple sub-objective function values, including a first sub-objective function value and a second sub-objective function value. The first sub-objective function value is obtained based on the energy market revenue, and the second sub-objective function value is obtained based on a preset wind curtailment penalty calculation formula, as described in Embodiment 1.

[0131] Example 3

[0132] Based on the above embodiments, the present invention also provides a terminal, such as... Figure 3 As shown, the terminal includes a processor 10 and a memory 20. Figure 3 Only some of the terminal components are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0133] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory. In other embodiments, the memory 20 may be an external storage device of the terminal, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Further, the memory 20 may include both internal and external storage devices. The memory 20 is used to store application software and various types of data installed on the terminal. The memory 20 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 20 stores a scheduling program 30 for offshore wind power storage participating in peak shaving ancillary services. This scheduling program 30 can be executed by the processor 10 to implement the scheduling method for offshore wind power storage participating in peak shaving ancillary services in this application.

[0134] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor, or other chip, used to run program code stored in the memory 20 or process data, such as executing the scheduling method for offshore wind power energy storage to participate in peak shaving auxiliary services.

[0135] In one embodiment, when the processor 10 executes the scheduling program 30 for offshore wind energy storage participating in peak shaving ancillary services in the memory 20, the following steps are performed:

[0136] Obtain initialization information, which includes wind power forecast information and electricity load demand information for the target date. Based on the initialization information, randomly generate initial solutions for multiple scheduling schemes. Add the initial solutions to the target set. Perform constraint processing on the solutions in the target set based on pre-set power grid constraints to obtain the updated target set.

[0137] For each solution in the target set, two solutions are randomly selected from its neighborhood, and a first solution is generated based on the two solutions.

[0138] The first solution is mutated multiple times based on the optimal sub-objective function value among the solutions in the current target set to generate a second solution. The target set is then updated based on the objective function value corresponding to the second solution and the objective function value corresponding to the solution in the current target set.

[0139] The process of re-exercising the step of randomly selecting two solutions in the neighborhood of each solution in the target set and performing gene recombination on the two solutions using an adaptive crossover operator is repeated until a preset number of iterations is reached.

[0140] After reaching the preset number of iterations, the scheduling scheme corresponding to the optimal solution is selected from the target set as the target scheduling scheme for the target date. The wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power and electricity consumption data are controlled according to the target scheduling scheme for each time period of the target date.

[0141] Each of the scheduling schemes includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date.

[0142] The objective function value includes multiple sub-objective function values, which include a first sub-objective function value and a second sub-objective function value. The first sub-objective function value is obtained based on the energy market revenue, and the second sub-objective function value is obtained based on a preset wind curtailment penalty calculation formula.

[0143] The sub-objective function values ​​also include: a third sub-objective function value and a fourth sub-objective function value;

[0144] The value of the third sub-objective function is obtained based on the preset loss cost of the energy storage system;

[0145] The value of the fourth sub-objective function is obtained based on the variance of the output power fluctuation of the wind turbine and the energy storage system.

[0146] The acquisition of initialization information includes:

[0147] Obtain wind forecast data for the target date;

[0148] The wind forecast data is input into a pre-trained neural network model to obtain the wind power prediction information output by the neural network model.

[0149] The power grid constraints include:

[0150] The wind power generation capacity does not exceed the wind power prediction information;

[0151] The charging and discharging power of the energy storage system shall not exceed the upper limit of the charging and discharging power of the energy storage system.

[0152] The fluctuations in wind power generation and the charging and discharging power of energy storage systems during different time periods shall not exceed the upper limit of fluctuation.

[0153] User consumption data within each time period does not exceed the pre-set power consumption limits for each user.

[0154] User consumption data within each time period satisfies the user load demand information.

[0155] The step of generating a first solution based on the two solutions includes:

[0156] An adaptive crossover operator is used to recombine the two solutions to generate the first solution.

[0157] The calculation formula for the adaptive crossover operator is:

[0158] PC i =a min +a max (i / N);

[0159] Where pc is the adaptive crossover operator in the i-th iteration, and a min For the preset minimum crossover probability, a max Where N is the preset maximum crossover probability, and N is the preset number of iterations.

[0160] The step of performing multiple mutations on the first solution based on the optimal sub-objective function value among the solutions in the current objective set to generate a second solution includes:

[0161] Based on the sub-objective function values ​​corresponding to all solutions in the current objective set, determine the optimal sub-objective function values;

[0162] Based on the optimal values ​​of each sub-objective function, the optimal distance corresponding to the first solution is determined by the first preset formula;

[0163] Based on the optimal distance corresponding to the first solution, the second solution is generated using a second preset formula during each mutation.

[0164] The first preset formula is:

[0165]

[0166] The optimal distance corresponds to the mutation result of the i-th first solution in the t-th iteration. f is the nth sub-objective function value corresponding to the tth mutation result of the i-th first solution.nmin This represents the optimal value of the nth sub-objective function in the current set of objectives.

[0167] The second preset formula is:

[0168]

[0169] stepsize is a variable asynchronous length. Let represent the result of the t-th mutation of the i-th first solution. Let represent the mutation result of the i-th first solution in the (t+1)-th iteration, where randon is a random number that follows a standard normal distribution, s(λ) is the random search path, and α is a scaling factor. This indicates a dot product operation. When the number of mutations reaches the preset number of mutations, the mutation stops, and the final mutation result is used as the second solution.

[0170] Wherein, updating the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the objective solution in the current target set includes:

[0171] For each of the second solutions, the following processing is performed:

[0172] The third solution is removed from the objective set according to the second solution, wherein each sub-objective function value of the third solution is worse than the corresponding sub-objective function value of the second solution;

[0173] If the fourth solution does not exist in the target set, the second solution is added to the target set, and each sub-objective function value in the fourth section is better than the sub-objective function value of the corresponding second solution.

[0174] Example 4

[0175] The present invention also provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the steps of the scheduling method for offshore wind power energy storage to participate in peak shaving ancillary services as described above.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A scheduling method for offshore wind power energy storage to participate in peak-shaving ancillary services, characterized in that, The method includes: Obtain initialization information, which includes wind power forecast information and electricity load demand information for the target date. Based on the initialization information, randomly generate initial solutions for multiple scheduling schemes. Add the initial solutions to the target set. Perform constraint processing on the solutions in the target set based on pre-set power grid constraints to obtain the updated target set. For each solution in the target set, two solutions are randomly selected from its neighborhood, and a first solution is generated based on the two solutions. The first solution is mutated multiple times based on the optimal sub-objective function value among the solutions in the current target set to generate a second solution. The target set is then updated based on the objective function value corresponding to the second solution and the objective function value corresponding to the solution in the current target set. The process of re-exercising the step of randomly selecting two solutions in the neighborhood of each solution in the target set and performing gene recombination on the two solutions using an adaptive crossover operator is repeated until a preset number of iterations is reached. After reaching the preset number of iterations, the scheduling scheme corresponding to the optimal solution is selected from the target set as the target scheduling scheme for the target date. The wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power and electricity consumption data are controlled according to the target scheduling scheme for each time period of the target date. Each of the scheduling schemes includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date. The objective function value includes multiple sub-objective function values, which include: a first sub-objective function value and a second sub-objective function value. The first sub-objective function value is obtained based on the energy market revenue, and the second sub-objective function value is obtained based on a preset wind curtailment penalty calculation formula. The step of performing multiple mutations on the first solution based on the optimal sub-objective function value among the solutions in the current objective set to generate a second solution includes: Based on the sub-objective function values ​​corresponding to all solutions in the current objective set, determine the optimal sub-objective function values; Based on the optimal values ​​of each sub-objective function, the optimal distance corresponding to the first solution is determined by the first preset formula; Based on the optimal distance corresponding to the first solution, the second solution is generated using a second preset formula during each mutation. The first preset formula is: ； The optimal distance corresponds to the t-th mutation result of the i-th first solution. Let be the nth sub-objective function value corresponding to the tth mutation result of the i-th first solution. This represents the optimal value of the nth sub-objective function in the current set of objectives. The second preset formula is: ； For variable asynchronous length, Let represent the result of the t-th mutation of the i-th first solution. Let represent the mutation result of the i-th first solution at the (t+1)-th iteration. The numbers are random numbers, which are random sequences that follow a standard normal distribution. For random search paths, As a scaling factor, This represents the dot product operation. When the number of mutations reaches the preset number of mutations, the mutation stops, and the final mutation result is used as the second solution.

2. The scheduling method for offshore wind power energy storage participating in peak-shaving ancillary services according to claim 1, characterized in that, The sub-objective function values ​​also include: the third sub-objective function value and the fourth sub-objective function value; The value of the third sub-objective function is obtained based on the preset loss cost of the energy storage system; The value of the fourth sub-objective function is obtained based on the variance of the output power fluctuation of the wind turbine and the energy storage system.

3. The scheduling method for offshore wind power energy storage participating in peak-shaving ancillary services according to claim 1, characterized in that, The acquisition of initialization information includes: Obtain wind forecast data for the target date; The wind forecast data is input into a pre-trained neural network model to obtain the wind power prediction information output by the neural network model.

4. The scheduling method for offshore wind power energy storage participating in peak-shaving ancillary services according to claim 1, characterized in that, The power grid constraints include: The wind power generation capacity does not exceed the wind power prediction information; The charging and discharging power of the energy storage system shall not exceed the upper limit of the charging and discharging power of the energy storage system. The fluctuations in wind power generation and the charging and discharging power of energy storage systems during different time periods shall not exceed the upper limit of fluctuation. User consumption data within each time period does not exceed the pre-set power consumption limits for each user. User consumption data for each time period meets the aforementioned electricity load demand information.

5. The scheduling method for offshore wind power energy storage participating in peak-shaving ancillary services according to claim 1, characterized in that, The step of generating a first solution based on the two solutions includes: An adaptive crossover operator is used to recombine the two solutions to generate the first solution. The calculation formula for the adaptive crossover operator is: ； in, The adaptive crossover operator in the i-th iteration, The minimum crossover probability is preset. Where N is the preset maximum crossover probability, and N is the preset number of iterations.

6. The scheduling method for offshore wind power energy storage participating in peak-shaving ancillary services according to claim 1, characterized in that, The step of updating the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the objective solution in the current target set includes: For each of the second solutions, the following processing is performed: The third solution is removed from the objective set according to the second solution, wherein each sub-objective function value of the third solution is worse than the corresponding sub-objective function value of the second solution; If the fourth solution does not exist in the target set, the second solution is added to the target set, and each sub-objective function value of the fourth solution is better than the corresponding sub-objective function value of the second solution.

7. A dispatching device for offshore wind power energy storage participating in peak-shaving ancillary services, characterized in that, The dispatching device for offshore wind power storage participating in peak shaving ancillary services is used to implement the dispatching method for offshore wind power storage participating in peak shaving ancillary services as described in any one of claims 1-6, including: An initialization module is used to acquire initialization information, including wind power forecast information and electricity load demand information for a target date. Based on the initialization information, multiple initial solutions for scheduling schemes are randomly generated, and the initial solutions are added to a target set. Based on pre-set constraints, the individuals in the target set are constrained to obtain an updated target set. The first mutation module is used to randomly select two solutions in the neighborhood of each target solution in the target set, and generate a first solution based on the two solutions. The second mutation module is used to perform multiple mutations on the first solution based on the optimal sub-objective function value among the solutions of the current target set to generate a second solution, and update the target set based on the objective function value corresponding to the second solution and the objective function value corresponding to the target solution in the current target set. An iterative module is used to re-execute the step of randomly selecting two solutions in the neighborhood of each target solution in the target set and performing gene recombination on the two solutions using an adaptive crossover operator until a preset number of iterations is reached. The scheduling module is used to select the scheduling scheme corresponding to the optimal solution from the target set as the target scheduling scheme for the target date after reaching a preset number of iterations, and to control the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power and electricity consumption data in each time period of the target date according to the target scheduling scheme. Each of the scheduling schemes includes the wind turbine power generation, energy storage system charging and discharging status, energy storage system charging and discharging power, and electricity consumption data for each time period within the target date. The objective function value includes multiple sub-objective function values, which include a first sub-objective function value and a second sub-objective function value. The first sub-objective function value is obtained based on the energy market revenue, and the second sub-objective function value is obtained based on a preset wind curtailment penalty calculation formula.

8. A terminal, characterized in that, The terminal includes: a processor and a computer-readable storage medium communicatively connected to the processor. The computer-readable storage medium is adapted to store multiple instructions, and the processor is adapted to call the instructions in the computer-readable storage medium to execute the steps of the scheduling method for offshore wind power energy storage participating in peak shaving ancillary services as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which can be executed by one or more processors to implement the steps of the scheduling method for offshore wind power energy storage participating in peak shaving ancillary services as described in any one of claims 1-6.

Images