Electrolyzer configuration processing method and apparatus based on variable wind-solar hybrid power generation

Abstract

The present description provides an electrolyzer configuration processing method and apparatus based on variable wind-solar hybrid power generation. The method comprises: first acquiring historical wind and solar data of a target area and constructing a wind-solar power output curve having good representativeness according to the historical wind and solar data of the target area; then, according to the wind-solar power output curve, determining and using a gradient step size to process the wind-solar power output curve, and extracting a base load component and a variable load component; and on the basis of a predefined matching rule and the gradient step size, performing optimization solving according to the base load component, the variable load component, and decision-influencing data to determine a target configuration solution that meets a requirement, wherein the predefined matching rule comprises: the base load component being preferentially matched with a first Nm³ / h-rated electrolyzer, and the variable load component being preferentially matched with a second Nm³ / h-rated electrolyzer; and arranging, according to the target configuration solution, a target electrolyzer bank for a target wind-solar hybrid power plant. This achieves an accurate and rational electrolyzer bank layout for a target wind-solar hybrid power plant.

Description

Method and apparatus for electrolyzer configuration processing based on wave-like wind-solar coupled power generation

[0001] This application claims priority to Chinese Patent Application No. 202510115556.3, filed on January 24, 2025, entitled “Method and Apparatus for Electrolyte Configuration Processing Based on Fluctuating Wind-Solar Coupling Power Generation”, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This specification belongs to the field of electrical data processing technology related to wind and solar new energy, and in particular to the electrolytic cell configuration processing method and device based on fluctuating wind and solar coupled power generation. Background Technology

[0003] With the promotion and development of wind and solar new energy technologies, many power plants have begun to use wind-solar coupling technology to generate electricity and convert the electricity generated by wind-solar coupling into hydrogen energy stored in electrolyzers. In the future, the hydrogen energy can be converted back into electricity as needed to provide it to downstream electricity users or feed it into the grid.

[0004] However, wind-solar coupled power generation is easily affected by natural conditions, leading to significant uncertainty in actual power generation. Based on existing methods, there are often problems with the unreasonable configuration and deployment of electrolyzers in power plants. For example, some power plants deploy electrolyzers of insufficient scale, resulting in the inability to fully convert the electricity generated by wind power into hydrogen energy for storage, leading to substantial power wastage. Conversely, some power plants deploy electrolyzers of excessive scale, increasing the construction and maintenance costs of the power plant and affecting its overall economic efficiency. Furthermore, this results in a large number of electrolyzers remaining idle for extended periods, wasting equipment resources.

[0005] There is currently no effective solution to the above problems. Summary of the Invention

[0006] This specification provides a method and apparatus for configuring and processing electrolyzers based on fluctuating wind-solar coupled power generation. It can be well adapted to wind-solar coupled power generation scenarios with fluctuations, effectively taking into account factors such as cost, benefits, and utilization rate, and accurately and reasonably realizing the deployment of electrolyzer groups for target wind-solar coupled power stations in the target area.

[0007] This specification provides a method for configuring and processing electrolyzers based on fluctuating wind-solar coupled power generation, including:

[0008] Obtain historical landscape data of the target area; wherein, the historical landscape data includes landscape data at multiple historical time points, and the landscape data at each historical time point also carries a timestamp corresponding to the historical time point;

[0009] Identify decision-making impact data for the target region;

[0010] According to the preset construction rules, the wind and solar power output curve is constructed using historical wind and solar data of the target area; wherein, the wind and solar power output curve is used to characterize the reference power of wind and solar coupled power generation at various time points within a year.

[0011] Based on the wind and solar power output curve, a matching gradient step size is determined; and using this gradient step size, the wind and solar power output curve is processed to separate the base load component and the fluctuating load component.

[0012] Based on preset matching rules and gradient step size, optimization is performed according to the base load component, fluctuating load component, and decision influence data to determine the target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolytic cell, and the fluctuating load component is preferentially matched with the second standard square electrolytic cell, wherein the standard square of the first standard square electrolytic cell is greater than that of the second standard square electrolytic cell.

[0013] According to the target configuration scheme, target electrolytic cells are deployed for the target wind-solar coupled power station to store the electrical energy generated by wind-solar coupled power generation.

[0014] In one embodiment, the decision-influencing data includes at least one of the following: electrolytic cell equipment and auxiliary equipment parameters, construction costs, operating costs, and operating revenue;

[0015] Accordingly, determine the decision-making impact data for the target area, including:

[0016] Obtain electricity price data and power grid rules for the target area;

[0017] By combining electricity price data and grid rules for the target area, decision-making impact data for the target area can be determined.

[0018] In one embodiment, a wind power output curve is constructed using historical wind and solar data of the target area according to preset construction rules, including:

[0019] Based on the timestamp, multiple power generation data groups are divided using historical landscape data of the target area; each power generation data group corresponds to a month, containing landscape data of multiple historical time points of that month in different years, as well as landscape data of multiple historical time points of different months in the same year that are adjacent to that month.

[0020] Based on multiple power generation data sets, multiple corresponding power generation data matrices are constructed; wherein each of the multiple power generation data matrices corresponds to a month.

[0021] Multiple power generation data matrices are processed using a pre-defined typical data generation model to obtain reference data sets for multiple wind and solar power generation in different months.

[0022] By splicing together and utilizing multiple sets of reference data on wind and solar power generation, the corresponding wind and solar power output curves are obtained.

[0023] In one embodiment, determining a matching gradient step size based on the wind and solar power output curve includes:

[0024] The wind and solar power output curve is subjected to graphic feature extraction to obtain the corresponding graphic features;

[0025] Based on the aforementioned graphic features, the fluctuation characteristics of the wind and solar power output curve are determined;

[0026] Based on the fluctuation characteristics, a matching gradient step size is determined.

[0027] In one embodiment, based on preset matching rules and gradient step size, optimization is performed according to the base load component, fluctuating load component, and decision impact data to determine a target configuration scheme that meets the requirements, including:

[0028] Construct a set of objective functions based on decision impact data; and construct objective constraints based on preset matching rules, base load components, and fluctuating load components.

[0029] Based on the preset processing rules, multiple computing nodes and the initial solutions for each computing node are determined.

[0030] Multiple computing nodes are invoked to perform multiple rounds of iterative solution for the objective function set based on the objective constraints and initial solution, so as to obtain the corresponding objective processing result;

[0031] Based on the target processing results, a target configuration scheme that meets the requirements is determined.

[0032] In one embodiment, multiple computing nodes are invoked to perform multiple rounds of iterative solutions to the objective function set based on the objective constraints and the initial solution, including:

[0033] Multiple computing nodes can be invoked for the current iteration in the following manner:

[0034] Obtain and filter out the solutions for the current round that meet the retention criteria based on the solutions from the previous round of each computing node;

[0035] The computing nodes that hold the retained solution for the current round are marked as first-class nodes, and the other computing nodes among the multiple computing nodes, excluding the first-class nodes, are marked as second-class nodes;

[0036] The retained solutions of the current round are processed by a preset encoding to obtain the corresponding encoded data; and based on the encoded data, the solutions are modified to obtain multiple modified codes.

[0037] The modification code is decoded to obtain multiple modification solutions for the current round; and the multiple modification solutions for the current round are assigned to the corresponding second-type nodes.

[0038] The first type of node and the second type of node are called to perform iterative solutions based on the local iterative solution rules and gradient step size, respectively using the retained solution and the modified solution of the current round, to obtain the solution of each computing node in the current round.

[0039] In one embodiment, after calling the first type of nodes and the second type of nodes to iteratively solve the problem according to the local iterative solution rules and gradient step size, respectively using the retained solution and the modified solution of the current round, to obtain the solution of each computing node for the current round, the method further includes:

[0040] Check whether the termination condition of the iterative solution is met;

[0041] If the termination condition of the iterative solution is met, the solution with the highest matching degree with the objective function set is selected from the solutions of the current round of multiple computing nodes and used as the objective processing result.

[0042] This specification also provides an electrolyzer configuration and processing device based on wave-dependent wind-solar coupled power generation, including:

[0043] The acquisition module is used to acquire historical landscape data of the target area; wherein, the historical landscape data includes landscape data at multiple historical time points, and the landscape data at each historical time point also carries a timestamp corresponding to the historical time point;

[0044] The determination module is used to determine the decision impact data for the target area;

[0045] The construction module is used to construct a wind and solar power output curve based on the historical wind and solar data of the target area according to the preset construction rules; wherein the wind and solar power output curve is used to characterize the reference power of wind and solar coupled power generation at various time points within a year.

[0046] The splitting module is used to determine the matching gradient step size based on the wind and solar power output curve; and to use the gradient step size to process the wind and solar power output curve and split the base load component and the fluctuating load component.

[0047] The solution module is used to perform optimization solutions based on preset matching rules and gradient step size, according to the base load component, fluctuating load component, and decision influence data, to determine the target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolytic cell, and the fluctuating load component is preferentially matched with the second standard square electrolytic cell, wherein the standard square of the first standard square electrolytic cell is greater than that of the second standard square electrolytic cell;

[0048] The deployment module is used to deploy target electrolytic cells for the target wind-solar coupled power station according to the target configuration scheme, in order to store the electrical energy generated by wind-solar coupled power generation.

[0049] This specification also provides a server, including a processor and a memory for storing processor-executable instructions, wherein the processor, when executing the instructions, implements the relevant steps of the electrolyzer configuration processing method based on fluctuating wind-solar coupled power generation.

[0050] This specification also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the steps of the electrolyzer configuration processing method based on fluctuating wind-solar coupled power generation.

[0051] Based on the electrolyzer configuration processing method and apparatus for fluctuating wind-solar coupled power generation provided in this specification, historical wind and solar data of the target area are first acquired; simultaneously, decision-making impact data for the target area is determined; according to preset construction rules, using the historical wind and solar data of the target area, a wind and solar power output curve with good stability is constructed, which can completely and representatively reflect the fluctuation of wind and solar coupled power generation throughout the year; then, based on the wind and solar power output curve, a matching gradient step size is determined; and using the gradient step size, the base load component and fluctuating load component are separated by processing the wind and solar power output curve; and based on the preset matching rules and gradient step size, optimization is performed according to the base load component, fluctuating load component, and decision-making impact data to determine a target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolyzer, and the fluctuating load component is preferentially matched with the second standard square electrolyzer, wherein the standard square of the first standard square electrolyzer is greater than that of the second standard square electrolyzer; according to the target configuration scheme, a target electrolyzer group for the target wind-solar coupled power station is deployed. This allows it to be well adapted to wind-solar coupled power generation scenarios with fluctuations, fully considers the fluctuating characteristics of wind and solar data in the target area throughout the year, effectively balances factors such as cost, revenue, and utilization rate, and accurately and reasonably realizes the deployment of electrolytic cell groups for the target wind-solar coupled power station in the target area, thereby improving the overall efficiency of the target wind-solar coupled power station. Attached Figure Description

[0052] To more clearly illustrate the embodiments of this specification, the accompanying drawings used in the embodiments will be briefly introduced below. The drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 is a flowchart illustrating an embodiment of the electrolyzer configuration processing method based on fluctuating wind-solar coupled power generation provided in this specification.

[0054] Figure 2 is a schematic diagram of an embodiment of the electrolyzer configuration processing method based on wave-driven wind-solar coupled power generation provided in this specification, applied in a scenario example.

[0055] Figure 3 is a schematic diagram of an embodiment of the electrolyzer configuration processing method based on wave-driven wind-solar coupled power generation provided in this specification, applied in a scenario example.

[0056] Figure 4 is a schematic diagram of an embodiment of the electrolyzer configuration processing method based on wave-coupled wind and solar power generation provided in this specification, applied in a scenario example.

[0057] Figure 5 is a schematic diagram of the structural composition of a server provided in one embodiment of this specification;

[0058] Figure 6 is a schematic diagram of the structure of an electrolyzer configuration processing device based on wave-driven wind-solar coupled power generation according to an embodiment of this specification;

[0059] Figure 7 is a schematic diagram of an embodiment of the electrolyzer configuration processing method based on wave-dependent wind-solar coupled power generation provided in this specification, applied in a scenario example. Detailed Implementation

[0060] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0061] It should be noted that in the embodiments of this specification, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0062] Referring to Figure 1, this specification provides an embodiment of an electrolyzer configuration and processing method based on fluctuating wind-solar coupled power generation. Specifically, this method is applied to the server side. In practical implementation, the method may include the following:

[0063] S101: Obtain historical landscape data of the target area; wherein, the historical landscape data includes landscape data at multiple historical time points, and the landscape data at each historical time point also carries a timestamp corresponding to the historical time point;

[0064] S102: Determine the decision impact data for the target area;

[0065] S103: Based on the preset construction rules, construct the wind and solar power output curve using historical wind and solar data of the target area; wherein, the wind and solar power output curve is used to characterize the reference power of wind and solar coupled power generation at various time points within a year.

[0066] S104: Determine the matching gradient step size based on the wind and solar power output curve; and use this gradient step size to process the wind and solar power output curve and separate the base load component and the fluctuating load component.

[0067] S105: Based on preset matching rules and gradient step size, optimize the solution according to the base load component, fluctuating load component, and decision influence data to determine the target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolytic cell, and the fluctuating load component is preferentially matched with the second standard square electrolytic cell, wherein the standard square of the first standard square electrolytic cell is greater than that of the second standard square electrolytic cell.

[0068] S106: According to the target configuration scheme, deploy a target electrolytic cell group for the target wind-solar coupled power station to store the electrical energy generated by wind-solar coupled power generation.

[0069] Specifically, the aforementioned target area can be understood as an area where a target wind-solar coupled power station is deployed.

[0070] The aforementioned wind-solar coupled power station can be understood as a new energy power station that uses wind-solar coupled power generation technology to generate electricity. Specifically, wind-solar coupled power generation refers to a power generation method that combines wind and solar energy. Based on this method, through the combined action of wind turbines and solar panels, the complementarity of wind and solar energy resources can be effectively utilized to provide a relatively stable and reliable power supply.

[0071] The historical landscape data for the target area mentioned above can be understood as the historical landscape data for the target area over the past few years, for example, the actual landscape data for each of the past 5 years. This landscape data may include wind data, sunshine data, etc.

[0072] Specifically, the aforementioned historical landscape data can include landscape data from multiple consecutive historical time points. The time interval between two adjacent historical time points can be one hour.

[0073] Specifically, the aforementioned historical landscape data can also carry timestamps corresponding to historical time points. Furthermore, the aforementioned historical landscape data can also include relevant information such as the year and month at the time of collection.

[0074] For example, wind and solar data at a historical point in time can be represented in the following form (F, G, T, Y, M). Here, F represents the wind data at that historical point in time, G represents the sunshine data at that historical point in time, T represents the timestamp corresponding to that historical point in time, Y represents the year, and M represents the month. Furthermore, the wind and solar data at the above historical point in time can also be represented in the following form (P, T, Y, M). Here, P represents the power generation at that historical point in time, determined based on the wind and sunshine data at that time, combined with the generator parameters of the target wind-solar coupled power station.

[0075] The aforementioned decision-making impact data can be understood as parameter data that affects the configuration and layout of electrolytic cells from multiple different perspectives, such as cost, benefit, and utilization rate.

[0076] The aforementioned wind and solar power output curves are specifically generated automatically based on artificial intelligence. These curves comprehensively represent representative and relatively stable reference power data for wind-solar coupled power generation at various points in time throughout the year. The time interval between two adjacent time points can be one hour. Specifically, the wind and solar power output curves can be plotted with time on the horizontal axis and wind-solar coupled power generation on the vertical axis.

[0077] Accordingly, in specific implementation, the aforementioned wind and solar power output curves can be used to finely divide the relatively stable base load components at different times and the relatively changing fluctuating load components between different time points using the gradient wavelength determined based on the fluctuation characteristics of the curves.

[0078] The aforementioned preset matching rules may specifically include: the base load component is preferentially matched with the first standard square electrolytic cell, and the fluctuating load component is preferentially matched with the second standard square electrolytic cell, wherein the standard square of the first standard square electrolytic cell is greater than that of the second standard square electrolytic cell.

[0079] The first standard cubic meter electrolytic cell mentioned above can also be called a large standard cubic meter electrolytic cell, specifically referring to an electrolytic cell with a standard cubic meter capacity greater than 1000. The second standard cubic meter electrolytic cell mentioned above can also be called a small standard cubic meter electrolytic cell, specifically referring to an electrolytic cell with a standard cubic meter capacity less than 1000.

[0080] Specifically, the standard volume of the first standard volume electrolytic cell is larger than that of the second standard volume electrolytic cell; correspondingly, the deployment cost and maintenance difficulty of the first standard volume electrolytic cell are higher than those of the second standard volume electrolytic cell, but the response speed of the first standard volume electrolytic cell is lower than that of the second standard volume electrolytic cell.

[0081] Based on the aforementioned preset matching rules, during the optimization process, the more expensive first standard cubic electrolyzer can be used to match the relatively stable extremely low load component, while the less expensive and more flexible second standard cubic electrolyzer can be used to match the relatively changing fluctuating load component.

[0082] Furthermore, the aforementioned second standard-size electrolyzer can include various standard-size electrolyzers, such as those with a standard-size of 500, 300, 200, and 50. Accordingly, in practical implementation, various second standard-size electrolyzers with different standard-sizes, along with the first standard-size electrolyzer, can be used in combination to better match the fluctuation characteristics of wind and solar data, thereby enabling a more reasonable and precise configuration and deployment.

[0083] Based on the above embodiments, historical wind and solar data of the target area can be acquired and utilized first to construct a wind and solar power output curve with good stability that can completely and representatively reflect the fluctuation of wind and solar coupled power generation throughout the year. Then, using the wind and solar power output curve and a gradient step size, the relatively stable base load component and the relatively changing fluctuating load component can be finely divided. Based on the preset matching rules and gradient step size, according to the base load component, fluctuating load component, and decision influence data, the target configuration scheme for the target wind and solar coupled power station in the target area can be determined efficiently and accurately through optimization. Then, based on the target configuration scheme, the electrolytic cell group of the target wind and solar coupled power station can be accurately and reasonably deployed.

[0084] In some embodiments, after acquiring historical landscape data of the target area, the method may preprocess the historical landscape data during implementation. This preprocessing includes: default value processing, anomaly detection processing, and noise filtering.

[0085] The anomaly detection process includes: for each historical time point's wind and solar data, detecting if the difference between the power generation (or wind and solar data) value at that historical time point and the average value of the data values ​​at a predetermined number (e.g., 10) of adjacent historical time points exceeds a predetermined threshold. If the difference exceeds this threshold, the wind and solar data at that historical time point is determined to be abnormal. In this case, data values ​​from the predetermined number of adjacent time points can be acquired and used to perform data fitting to obtain a fitted curve; then, based on the fitted curve, the fitted value for that historical time point is determined; and the fitted value is used to replace the original data value for that historical time point.

[0086] In some embodiments, the decision-influence data may specifically include at least one of the following: parameters of the electrolytic cell equipment and auxiliary equipment, construction costs, operating costs, operating revenue, etc.

[0087] Accordingly, the aforementioned determination of decision-making impact data for the target area, in specific implementation, may include:

[0088] S1: Obtain electricity price data and power grid rules for the target area;

[0089] S2: By combining electricity price data and grid rules for the target area, determine the decision-making impact data for the target area.

[0090] Specifically, the parameters of the aforementioned electrolytic cell equipment and auxiliary equipment may include: the price of each standard cubic meter electrolytic cell (including the first standard cubic meter electrolytic cell and the second standard cubic meter electrolytic cell), as well as the price and floor space of the corresponding auxiliary equipment.

[0091] The aforementioned construction costs may specifically include one or more of the following: the cost of wind-solar coupled power generation equipment, the cost of hydrogen production equipment, the cost of energy storage equipment, and transmission costs, etc. Furthermore, the aforementioned construction costs may also include other expenses such as land acquisition fees and preliminary fees payable during the construction period.

[0092] The aforementioned operating costs may include one or more of the following: personnel costs, material costs, repair costs, insurance premiums, construction period interest, periodic rental payments during the operating period, energy storage core replacement costs, electrolyzer maintenance costs, depreciation, and safety production costs, other manufacturing costs, other management costs, water and electricity costs, catalyst costs, etc., incurred during hydrogen production.

[0093] The revenue generated during the aforementioned operating period may specifically include: hydrogen price data, hydrogen sales volume, electricity sales volume, electricity price data, etc.

[0094] The aforementioned grid rules may specifically include grid-connected rules or off-grid rules. The grid-connected rules specify the timing of energy feeding into the grid, as well as the required amount of electricity to be fed into the grid at different times, related reward data, and penalty data. The off-grid rules specify the timing for the grid to refuse energy feeding, as well as the triggering conditions for refusing energy feeding.

[0095] In practice, electricity price data and power grid rules for the target area can be combined to precisely determine the decision impact data for each time point in the target area.

[0096] Based on the above embodiments, relatively accurate and effective, and time-point-based fine-grained decision impact data can be obtained, so that a better objective function set and objective constraints can be constructed based on the above decision impact data.

[0097] In some embodiments, referring to Figure 2, the above-mentioned construction of the wind and solar power output curve using historical wind and solar data of the target area according to preset construction rules may include the following in specific implementations:

[0098] S1: Based on the timestamp, use the historical landscape data of the target area to divide it into multiple power generation data groups; among them, one power generation data group corresponds to one month, which contains landscape data of multiple historical time points of that month in different years, as well as landscape data of multiple historical time points of different months in the same year that are adjacent to that month.

[0099] S2: Based on multiple power generation data sets, construct corresponding multiple power generation data matrices; wherein, each of the multiple power generation data matrices corresponds to a month;

[0100] S3: Use a preset typical data generation model to process multiple power generation data matrices to obtain reference data sets for multiple wind and solar power generation in different months;

[0101] S4: By splicing together and utilizing multiple sets of reference data on wind and solar power generation, the corresponding wind and solar power output curves are obtained.

[0102] In practice, based on the timestamp, combined with year and month information, wind and solar data (e.g., power generation) from multiple historical time points within the same month but from different years can be filtered out, along with wind and solar data from multiple historical time points within adjacent months of the same year (e.g., the previous month and the next month). These are then combined to obtain power generation data groups for the corresponding month. Following this method, 12 power generation data groups can be obtained, each corresponding to one of the 12 months.

[0103] For each power generation data set, a power generation data matrix corresponding to the month is constructed using wind and solar data from multiple historical time points within that data set, following the corresponding matrix construction rules. In this power generation data matrix, each row corresponds to a year, containing the wind and solar data values ​​for that month and multiple historical time points of adjacent months, arranged in sequence. Each row also corresponds to a timestamp, containing the wind and solar data values ​​for different years at the same historical time point corresponding to that timestamp.

[0104] Each power generation data matrix is ​​processed using a pre-trained, pre-defined typical data generation model to obtain multiple reference data sets for wind and solar power generation corresponding to different months. Each reference data set for wind and solar power generation contains representative, relatively stable, and highly valuable reference data values ​​for power generation at various time points within the corresponding month.

[0105] Using the above method and a pre-set typical data generation model, with months as the group unit of time nodes, the wind and light data at various time points in each month are studied and sorted from multiple angles based on the vertical dimension (adjacent months in the same year) and the horizontal dimension (the same months in different years). This effectively reduces the bias caused by the volatility of wind and light data, and obtains multiple reference data groups that are relatively stable and have good reference value and representativeness.

[0106] Furthermore, we can first use the aforementioned multiple sets of reference data on wind and solar power generation to construct curve segments for each individual month; then, according to the order of the months, we can sequentially splice multiple curve segments; and smooth the connection positions of adjacent curve segments to obtain the wind and solar power output curve that corresponds to the whole year.

[0107] Specifically, the aforementioned pre-set typical data generation model can be understood as follows: it is a model that uses big data samples to train through deep learning, which can take a month as the overall field of view and extract and integrate the volatility feature vectors of the data from two different dimensions, vertical and horizontal, for a single time point; then, based on the integrated volatility feature vector of the data at that time point, and in combination with the influence relationship of data from other time points adjacent to that time point in the vertical and horizontal directions, it determines and outputs the reference power of typical wind and solar power generation at that time point.

[0108] Based on the above embodiments, by using a preset typical data generation model to process the power generation data matrix of the corresponding structure, the fluctuation characteristics of wind and solar data can be fully considered, and wind and solar power output curves with good representativeness and high reference value can be constructed efficiently and accurately.

[0109] In some embodiments, the matching gradient step size is determined based on the wind and solar power output curve. In specific implementations, this may include the following:

[0110] S1: Extract graphic features from the wind and solar power output curve to obtain the corresponding graphic features;

[0111] S2: Based on the aforementioned graphic features, determine the fluctuation characteristics of the wind and solar power output curve;

[0112] S3: Determine the matching gradient step size based on the fluctuation characteristics.

[0113] In practice, multiple adjacent graphic features can be grouped into a graphic feature group; then, adjacent graphic feature groups can be combined and used to determine multiple fluctuation difference features through feature difference calculation; and then, based on multiple fluctuation difference features, fluctuation features for the wind and solar power output curve can be determined.

[0114] Based on the above embodiments, the fluctuation characteristics of the wind and solar power output curves can be fully explored and utilized to construct a gradient step size that is effective for the wind and solar power output curves.

[0115] In practical implementation, a gradient step size can be used to interpolate the wind and solar power output curve to obtain a trapezoidal composite diagram that approximates the curve. From this trapezoidal composite diagram, the graphical region of time points with heights less than a threshold value is segmented as the base load component (denoted as LB). The remaining graphical region within the graph enclosed by the wind and solar power output curve and the horizontal axis, excluding the base load component, is designated as the fluctuating load component (denoted as LF). This fully considers the fluctuating characteristics of wind and solar data and accurately identifies the base load component and fluctuating load component, which have higher application value.

[0116] In some embodiments, referring to Figure 3, the above-mentioned optimization solution based on the base load component, fluctuating load component, and decision impact data determines a target configuration scheme that meets the requirements. In specific implementation, this may include the following:

[0117] S1: Construct a set of objective functions based on decision impact data; and construct objective constraints based on preset matching rules, base load components, and fluctuating load components;

[0118] S2: Based on the preset processing rules, determine multiple computing nodes and the initial solution for each computing node;

[0119] S3: Call multiple computing nodes to perform multiple rounds of iterative solution on the objective function set based on the objective constraints and initial solution, and obtain the corresponding objective processing result;

[0120] S4: Based on the target processing results, determine the target configuration scheme that meets the requirements.

[0121] In practical implementation, the following functions can be constructed based on the decision-making impact data: an economic benefit maximization function, an idle electrolyzer volume minimization function, and a waste power minimization function. These functions are then combined to construct a set of objective functions. Simultaneously, based on preset matching rules, grid rules, and the performance upper limits of the generator parameters of the target wind-solar coupled power station, corresponding objective constraints are constructed.

[0122] The decision variables to be optimized in the above objective function set may include at least the number of first standard-square electrolytic cells and the number of second standard-square electrolytic cells with different standard-square values.

[0123] Furthermore, the aforementioned decision variables may also include: the connection relationship between the first standard cubic meter electrolyzer and second standard cubic meter electrolyzers with different standard cubic meter values, and the connection relationship between second standard cubic meter electrolyzers with different standard cubic meter values. Based on the above connection relationships, the priority order for using the second standard cubic meter electrolyzer can be determined when the hydrogen energy storage capacity of the first standard cubic meter electrolyzer reaches its full value; and the priority order for using the first standard cubic meter electrolyzer and the second standard cubic meter electrolyzer when it is necessary to convert and provide electrical energy to external sources.

[0124] The aforementioned pre-defined processing rules can specifically be solution rules that integrate multiple optimization algorithms (e.g., genetic algorithm, simulated annealing algorithm, etc.) and are adapted to wind-solar coupled power generation scenarios with fluctuations.

[0125] In practice, the overall computational load can be estimated based on the preset processing rules and the data scale of the fluctuating load components; and an appropriate number of computing nodes can be determined based on the estimated computational load.

[0126] Specifically, the aforementioned computing nodes can be relatively independent computing nodes (e.g., computer nodes) within a distributed cluster. Each computing node can be viewed as an individual in a population, deploying corresponding local iterative solution rules and connected to different computing nodes. The aforementioned distributed cluster can be understood as a system composed of multiple computer nodes that communicate and collaborate via a network to jointly complete a task or provide a service, offering advantages such as high availability, high performance, and scalability. In this way, the performance advantages of distributed systems can be leveraged to efficiently complete relevant solutions.

[0127] In practice, multiple initial solutions can be randomly generated and assigned to each computing node. Then, according to the preset processing rules, multiple computing nodes are called to perform multiple rounds of iterative solutions to the optimization problem of the objective function set based on the objective constraints and the initial solutions, so as to obtain the target configuration scheme that meets the requirements.

[0128] Based on the above embodiments, multiple computing nodes can be invoked to solve the problem through multiple rounds of iteration according to preset processing rules, which can avoid getting trapped in local optima and efficiently and accurately find the target configuration scheme that meets the requirements.

[0129] In some embodiments, the above-mentioned invocation of multiple computing nodes performs multiple rounds of iterative solution for the objective function set based on the objective constraints and the initial solution. Specifically, referring to Figure 4, multiple computing nodes can be invoked to perform the current round of iterative solution in the following manner:

[0130] S1: Obtain and filter out the solutions for the current round that meet the retention conditions based on the solutions from the previous round of each computing node;

[0131] S2: Mark the computing nodes that hold the reserved solution of the current round as first-class nodes, and mark the other computing nodes other than the first-class nodes as second-class nodes;

[0132] S3: Perform preset encoding processing on the retained solution of the current round to obtain the corresponding encoded data; and perform modification processing based on the encoded data to obtain multiple modified codes;

[0133] S4: Decode the modification code to obtain multiple modification solutions for the current round; and assign the multiple modification solutions for the current round to the corresponding second-type nodes;

[0134] S5: Call the first type of node and the second type of node to perform iterative solutions based on the local iterative solution rules and gradient step size, respectively using the retained solution of the current round and the modified solution of the current round, to obtain the solution of each computing node in the current round.

[0135] In practice, the solutions from the previous round of each computing node can be encoded according to the preset encoding rules to obtain the encoded data of the solutions from the previous round. Then, the selection operator constructed based on the pre-configured loss function of the objective function group can be used to process the encoded data of the solutions from the previous round to determine the selected encoded data. The solution corresponding to the selected encoded data is determined as the retained solution of the current round that meets the retention conditions.

[0136] The aforementioned preset encoding rule can be a floating-point encoding rule. Based on the floating-point encoding rule, data values ​​can be represented by a floating-point number within a certain range. This is suitable for integration with other types of optimization algorithms and helps to improve the complexity of the solution calculation, enabling a relatively larger solution space search and avoiding getting trapped in local optima.

[0137] In practice, pre-configured crossover and mutation operators can be used to process the encoded data of the retained solution in the current round to achieve the transformation process and obtain multiple corresponding transformation codes. Then, according to the preset encoding rules, the above transformation codes are decoded to obtain the corresponding transformation solution of the current round. The transformation solution of the current round is then assigned to the corresponding second type node to replace the original solution of the second type node in the previous round.

[0138] Alternatively, the solution of the second type of node in the previous round can be encoded according to the preset encoding rules to obtain the initial encoding; then, using the multiple encoded data obtained based on the retained solution of the current round, combined with the above initial encoding, the corresponding optimization vector can be calculated; then, the optimization vector can be used to optimize and transform the initial encoding of the second type of node to obtain the modified encoding of each second type of node.

[0139] Then, the first type of node is called to perform multiple iterations based on the local iterative solution rules and gradient step size, using the retained solution of the current round; at the same time, the second type of node is called to perform multiple iterations based on the local iterative solution rules and gradient step size, using the modified solution of the current round, to complete the iterative solution of the current round and obtain the solution of each computing node for the current round.

[0140] Specifically, the aforementioned local iterative solution rules can be solution rules that are modified and designed based on optimization algorithms.

[0141] Specifically, different computing nodes can deploy different local iterative solution rules. For example, computing nodes with node numbers ending in 1, 3, and 5 can deploy local iterative solution rules based on simulated annealing; computing nodes with node numbers ending in 2, 4, and 6 can deploy local iterative solution rules based on conjugate gradient algorithm; and computing nodes with node numbers ending in 0, 7, 9, and 8 can deploy local iterative solution rules based on quasi-Newton algorithm.

[0142] In this way, the advantages of different optimization algorithms can be fully utilized in each round of iterative solution, effectively increasing the search mechanism and expanding the search space, so as to search for the optimal solution more comprehensively and precisely.

[0143] Based on the above embodiments, according to the preset processing rules, by integrating and using multiple different optimization algorithms to complete each round of iterative solution, the efficiency and accuracy of the solution calculation can be effectively improved, and the overall processing time can be shortened.

[0144] In some embodiments, after calling the first type of nodes and the second type of nodes to iteratively solve the problem using the retained solution and the modified solution of the current round, respectively, according to the local iterative solution rules and gradient step size, to obtain the solution of each computing node for the current round, the method may further include the following:

[0145] S1: Check if the termination condition for iterative solution is met;

[0146] S2: If the termination condition of the iterative solution is met, select the solution with the highest matching degree with the objective function group from the solutions of the current round of multiple computing nodes, and use it as the objective processing result.

[0147] In practice, it can be detected whether the number of rounds in the current round has reached the specified number of rounds; when it is determined that the specified number of rounds has been reached, the deviation value between the solutions of multiple computing nodes in the current round is calculated, and it is detected whether the deviation value is less than the preset deviation threshold; when it is determined that it is less than the preset deviation threshold, it can be determined that a stable solution has been obtained, and thus it can be determined that the termination condition of iterative solution is met.

[0148] If the termination condition of the iterative solution is met, the solutions of the current round of multiple computing nodes can be substituted into the objective function group to obtain the joint evaluation result based on the objective function value. According to the joint evaluation result, the solution with the best effect based on multiple factors such as cost, benefit, and utilization rate is selected from the solutions of the current round of multiple computing nodes. This solution is the one with the highest matching degree with the objective function group and is used as the objective processing result.

[0149] Specifically, for example, the solutions of multiple computing nodes in the current round can be substituted into the objective function set, and the IRR economic benefit index can be calculated using the objective function set; the IRR economic benefit index can then be used as the joint evaluation result.

[0150] Conversely, if the termination condition for iterative solution is not met, the above process can be repeated to perform the next round of iterative solution until the target result is obtained.

[0151] Based on the above embodiments, the target processing result that meets the requirements can be accurately obtained through multiple rounds of iterative solution.

[0152] In some embodiments, after deploying the target electrolyzer group for the target wind-solar coupled power station according to the target configuration scheme, the method may further include the following in its specific implementation:

[0153] S1: Receive data on electricity prices and changes to grid rules for the target area;

[0154] S2: Based on the changed data, reconstruct the objective function set and objective constraints;

[0155] S3: Based on the preset processing rules, determine multiple computing nodes; and based on the previously determined target processing results, determine the initial solution for each computing node;

[0156] S4: Call multiple computing nodes to keep the number of configurations for the first standard electrolytic cell constant according to the target processing results; and perform multiple rounds of iterative solution for the objective function set based on the target constraints and initial solution to obtain the configuration optimization scheme;

[0157] S5: Based on the configuration optimization scheme, keep the previously deployed first standard electrolyzer unchanged, optimize and improve the target electrolyzer group to obtain an optimized target electrolyzer group to adapt to the latest wind-solar coupled power generation scenario.

[0158] Based on the above embodiments, the optimization and upgrading of the target electrolytic cell group can be achieved efficiently at a relatively low cost to meet the latest scenario requirements.

[0159] In some embodiments, after deploying the target electrolyzer group for the target wind-solar coupled power station according to the target configuration scheme, the method may further include: when the energy storage capacity of the first standard-sized electrolyzer reaches its full value, determining the start-up priority of the second standard-sized electrolyzer based on the connection relationship between the first and second standard-sized electrolyzers in the target configuration scheme; and starting the second standard-sized electrolyzers sequentially according to the start-up priority. This allows for the effective storage of newly generated wind-solar coupled power while better balancing overall costs to achieve better economic benefits.

[0160] As can be seen from the above, the electrolyzer configuration processing method based on fluctuating wind-solar coupled power generation provided in this specification first acquires historical wind and solar data of the target area; simultaneously, it determines the decision-influence data for the target area; according to preset construction rules, it uses the historical wind and solar data of the target area to construct a wind and solar power output curve with good stability that can completely and representatively reflect the fluctuation of wind and solar coupled power generation throughout the year; then, based on the wind and solar power output curve, it determines a matching gradient step size; and uses this gradient step size to process the wind and solar power output curve, separating the base load component and the fluctuating load component; and based on the preset matching rules and gradient step size, it performs optimization and solution based on the base load component, the fluctuating load component, and the decision-influence data to determine a target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolyzer, and the fluctuating load component is preferentially matched with the second standard square electrolyzer, wherein the standard square of the first standard square electrolyzer is greater than that of the second standard square electrolyzer; according to the target configuration scheme, the target electrolyzer group for the target wind-solar coupled power station is deployed. This allows it to be well adapted to wind-solar coupled power generation scenarios with fluctuations, fully consider the fluctuation characteristics of wind and solar data in the target area throughout the year, effectively balance factors such as cost, revenue, and utilization rate, and accurately and reasonably realize the deployment of electrolytic cell groups for target wind-solar coupled power stations in the target area.

[0161] This specification provides a server, as shown in Figure 5. The server includes a network communication port 501, a processor 502, and a memory 503. These components are connected via internal cables to enable data exchange between them.

[0162] Specifically, the network communication port 501 can be used to acquire historical landscape data of the target area; wherein, the historical landscape data includes landscape data at multiple historical time points, and the landscape data at each historical time point also carries a timestamp corresponding to that historical time point.

[0163] The processor 502 can specifically be used to determine decision impact data for a target area; construct a wind and solar power output curve using historical wind and solar data of the target area according to preset construction rules; wherein the wind and solar power output curve is used to characterize the reference power of wind and solar coupled power generation at various time points within a year; determine a matching gradient step size based on the wind and solar power output curve; and use the gradient step size to process the wind and solar power output curve, separating the base load component and the fluctuating load component; based on the preset matching rules and gradient step size, perform optimization solutions according to the base load component, the fluctuating load component, and the decision impact data to determine a target configuration scheme that meets the requirements; wherein the preset matching rules include: the base load component is preferentially matched with a first standard square electrolyzer, and the fluctuating load component is preferentially matched with a second standard square electrolyzer, wherein the standard square of the first standard square electrolyzer is greater than that of the second standard square electrolyzer; and deploy a target electrolyzer group for the target wind and solar coupled power station according to the target configuration scheme to store the electrical energy generated by wind and solar coupled power generation.

[0164] The memory 503 can be used to store corresponding instruction programs, as well as decision impact data, wind and solar power output curves, base load components, fluctuating load components, objective function sets, objective constraints, and other related data.

[0165] Based on the above method, the relevant structural performance of the server can be effectively utilized to improve the data processing speed of electronic devices and efficiently realize the data processing of electrolyzers based on fluctuating wind-solar coupled power generation.

[0166] In this embodiment, the network communication port 501 can be a virtual port bound to different communication protocols, thereby enabling the sending or receiving of different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0167] In this embodiment, the processor 502 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. This specification is not limiting.

[0168] In this embodiment, the memory 503 may include multiple layers. In a digital system, anything that can store binary data can be a memory. In an integrated circuit, a circuit with storage function but no physical form is also called a memory, such as RAM, FIFO, etc. In a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0169] This specification also provides a computer-readable storage medium based on the above-described electrolyzer configuration processing method for wave-coupled wind and solar power generation. The computer-readable storage medium stores computer program instructions that, when executed, implement the following steps: acquiring historical wind and solar data for a target area; wherein the historical wind and solar data includes wind and solar data from multiple historical time points, and the wind and solar data at each historical time point also carries a timestamp corresponding to that historical time point; determining decision-making impact data for the target area; and constructing a wind and solar power output curve using the historical wind and solar data of the target area according to preset construction rules; wherein the wind and solar power output curve is used to characterize the reference wind and solar power generation at various time points within a year. Power; based on the wind and solar power output curves, a matching gradient step size is determined; and using this gradient step size, the wind and solar power output curves are processed to separate the base load component and the fluctuating load component; based on preset matching rules and gradient step size, optimization is performed according to the base load component, fluctuating load component, and decision influence data to determine a target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolyzer, and the fluctuating load component is preferentially matched with the second standard square electrolyzer, wherein the standard square of the first standard square electrolyzer is greater than that of the second standard square electrolyzer; according to the target configuration scheme, a target electrolyzer group for the target wind and solar coupled power station is deployed to store the electrical energy generated by wind and solar coupled power generation.

[0170] In this embodiment, the storage medium includes, but is not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), cache, hard disk drive (HDD), or memory card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured according to the standards specified in the communication protocol for network connection communication.

[0171] In this embodiment, the specific functions and effects implemented by the program instructions stored in the computer-readable storage medium can be explained in comparison with other embodiments, and will not be repeated here.

[0172] This specification also provides a computer program product, comprising at least a computer program, which, when executed by a processor, implements the following method steps: acquiring historical wind and solar data of a target area; wherein the historical wind and solar data includes wind and solar data from multiple historical time points, and the wind and solar data at each historical time point also carries a timestamp corresponding to that historical time point; determining decision-making impact data for the target area; constructing a wind and solar power output curve using the historical wind and solar data of the target area according to preset construction rules; wherein the wind and solar power output curve is used to characterize the reference power of wind and solar coupled power generation at various time points within a year; and determining a matching [data / method] based on the wind and solar power output curve. A gradient step size is used to process the wind and solar power output curves, separating the base load component and the fluctuating load component. Based on preset matching rules and the gradient step size, optimization is performed according to the base load component, fluctuating load component, and decision influence data to determine a target configuration scheme that meets the requirements. The preset matching rules include: the base load component is preferentially matched with the first standard square cell, and the fluctuating load component is preferentially matched with the second standard square cell, wherein the standard square of the first standard square cell is greater than that of the second standard square cell. According to the target configuration scheme, a target electrolytic cell group is deployed for the target wind-solar coupled power station to store the electrical energy generated by wind-solar coupled power generation.

[0173] Referring to Figure 6, this specification also provides an electrolyzer configuration and processing device based on fluctuating wind-solar coupled power generation. This device may specifically include the following structural modules:

[0174] The acquisition module 601 can be used to acquire historical landscape data of a target area; wherein, the historical landscape data includes landscape data at multiple historical time points, and the landscape data at each historical time point also carries a timestamp corresponding to the historical time point;

[0175] Module 602 is specifically used to determine decision impact data for a target area;

[0176] The construction module 603 can be specifically used to construct a wind and solar power output curve based on the historical wind and solar data of the target area according to the preset construction rules; wherein, the wind and solar power output curve is used to characterize the reference power of wind and solar coupled power generation at various time points within a year.

[0177] The splitting module 604 can be used to determine the matching gradient step size based on the wind and solar power output curve; and use the gradient step size to process the wind and solar power output curve and split the base load component and the fluctuating load component.

[0178] The solution module 605 can be used to perform optimization solutions based on preset matching rules and gradient step size, according to the base load component, fluctuating load component, and decision influence data, to determine the target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolytic cell, and the fluctuating load component is preferentially matched with the second standard square electrolytic cell, wherein the standard square of the first standard square electrolytic cell is greater than that of the second standard square electrolytic cell;

[0179] The deployment module 606 can be used to deploy a target electrolytic cell group for the target wind-solar coupled power station according to the target configuration scheme, in order to store the electrical energy generated by wind-solar coupled power generation.

[0180] In some embodiments, the decision-influencing data includes at least one of the following: electrolytic cell equipment and auxiliary equipment parameters, construction costs, operating costs, and operating revenue;

[0181] Accordingly, when the aforementioned determining module 602 is specifically implemented, the decision impact data for the target area can be determined in the following manner: obtain the electricity price data and power grid rules of the target area; combine the electricity price data and power grid rules of the target area to determine the decision impact data for the target area.

[0182] In some embodiments, when the above-mentioned construction module 603 is specifically implemented, it can construct a wind and solar power output curve using historical wind and solar data of the target area according to a preset construction rule in the following manner: Based on the timestamp, the historical wind and solar data of the target area is divided into multiple power generation data groups; wherein, each power generation data group corresponds to a month, containing wind and solar data of multiple historical time points of that month in different years, as well as wind and solar data of multiple historical time points of different months adjacent to that month in the same year; based on the multiple power generation data groups, multiple corresponding power generation data matrices are constructed; wherein, each of the multiple power generation data matrices corresponds to a month; multiple power generation data matrices are processed using a preset typical data generation model to obtain multiple reference data groups of wind and solar power generation corresponding to different months; and the reference data groups of multiple wind and solar power generation are spliced ​​together and used to obtain the corresponding wind and solar power output curve.

[0183] In some embodiments, when the above-mentioned splitting module 604 is specifically implemented, the matching gradient step size can be determined according to the wind and solar power output curve in the following manner: extracting graphic features from the wind and solar power output curve to obtain corresponding graphic features; determining the fluctuation features of the wind and solar power output curve according to the graphic features; and determining the matching gradient step size according to the fluctuation features.

[0184] In some embodiments, when the solution module 605 is specifically implemented, it can perform optimization solutions based on preset matching rules and gradient step size, according to the base load component, fluctuating load component, and decision influence data, to determine a target configuration scheme that meets the requirements: constructing an objective function set based on the decision influence data; constructing target constraints based on preset matching rules, base load component, and fluctuating load component; determining multiple computing nodes and the initial solutions of each computing node according to preset processing rules; calling multiple computing nodes to perform multiple rounds of iterative solutions on the objective function set based on the target constraints and initial solutions to obtain the corresponding target processing results; and determining a target configuration scheme that meets the requirements based on the target processing results.

[0185] In some embodiments, when the solution module 605 is specifically implemented, it can call multiple computing nodes to perform iterative solution for the current round in the following manner: obtain and filter out the retained solutions for the current round that meet the retention conditions based on the solutions of each computing node in the previous round; mark the computing nodes holding the retained solutions for the current round as first-type nodes, and mark the other computing nodes among the multiple computing nodes other than the first-type nodes as second-type nodes; perform preset encoding processing on the retained solutions for the current round to obtain corresponding encoded data; and perform modification processing based on the encoded data to obtain multiple modified codes; perform decoding processing on the modified codes to obtain multiple modified solutions for the current round; and assign the multiple modified solutions for the current round to the corresponding second-type nodes; call the first-type nodes and the second-type nodes to perform iterative solution using the retained solutions and modified solutions for the current round respectively, according to the local iterative solution rules and gradient step size, to obtain the solution for the current round of each computing node.

[0186] In some embodiments, after calling the first type of node and the second type of node to iteratively solve the current round using the retained solution and the modified solution of the current round respectively according to the local iterative solution rules and gradient step size, and obtaining the solution of the current round of each computing node, the solution module 605 can also be used to detect whether the iterative solution termination condition is met; if it is determined that the iterative solution termination condition is met, the solution with the highest matching degree with the objective function group is selected from the solutions of the current round of multiple computing nodes as the objective processing result.

[0187] It should be noted that the units, devices, or modules described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. For ease of description, the above devices are described by dividing them into various modules according to their functions. Of course, in implementing this specification, the functions of each module can be implemented in one or more software and / or hardware, or the module that implements the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection between the devices or units shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0188] As can be seen from the above, the electrolytic cell configuration and processing device based on fluctuating wind-solar coupled power generation provided in the embodiments of this specification can be well adapted to wind-solar coupled power generation scenarios with fluctuations, fully consider the fluctuation characteristics of wind and solar data throughout the year in the target area, effectively take into account factors such as cost, benefits, and utilization rate, and accurately and reasonably realize the deployment of electrolytic cell groups for the target wind-solar coupled power station in the target area.

[0189] In a specific scenario example, the electrolyzer configuration processing method based on fluctuating wind-solar coupled power generation provided in this specification can be applied to achieve gradient matching of different standard square electrolyzers under fluctuating power sources. The specific implementation process may include the following.

[0190] Given that existing methods are often designed for stable power sources, most employ electrolyzers with a standardized form factor and corresponding equipment. However, in fluctuating power environments, using electrolyzers with a standardized form factor often leads to two problems: first, insufficient electrolyzer capacity, resulting in power curtailment during periods of abundant renewable energy generation; second, over-saturation of electrolyzers, preventing the full utilization of their hydrogen production capacity. Both of these scenarios prevent the project from achieving optimal overall economic benefits.

[0191] To address the aforementioned issues, the applicant, through creative thinking, proposed a method for mixing and matching different standard-size electrolyzers under fluctuating power sources. This method fully considers the resource endowment of the power source and pairs it with electrolyzers whose output curves are more closely aligned with it. The goal is to respond to renewable energy fluctuations by utilizing smaller standard-size electrolyzers (e.g., a second standard-size electrolyzer) as much as possible, while ensuring that larger standard-size electrolyzers (e.g., the first standard-size electrolyzer) maintain stable operating conditions. This aims to reduce overall project investment and increase the lifespan of larger standard-size electrolyzers. See Figure 7 for details, which includes the following steps.

[0192] S1: Obtain real-time landscape data (e.g., historical landscape data of the target area) for 8760 years in the target area and preprocess it.

[0193] Preferably, the real data of wind and solar 8760 mentioned in step S1 includes timestamps and corresponding time output (e.g., power generation of wind-solar coupled power generation).

[0194] Preferably, the preprocessing of the real data of wind and solar power 8760 includes the processing of the default values ​​of the timestamp and the corresponding power field, and the handling of anomalies, to obtain the processed power field P, the timestamp field T, the corresponding year Y, and the corresponding month M. Finally, the processed real sample data structure of 8760 is represented as D1 = (P, T, Y, M).

[0195] S2: Obtain local electricity prices and related policies (e.g., electricity price data and grid rules for the target area).

[0196] Preferably, the data obtained in step S2 includes local electricity prices, grid connection and off-grid policies such as requirements for grid connection or off-grid operation; permitted grid connection and off-grid electricity volumes; and mandatory requirements for project-supporting energy storage capacity.

[0197] S3: Obtain the current prices and floor space of each bidder's electrolytic cell and corresponding auxiliary equipment, as well as other data required for evaluation calculations (e.g., decision impact data), and preprocess them.

[0198] Preferably, the data in step S3 includes four main categories: electrolyzer equipment and auxiliary machine parameters, construction period investment, and operation period revenue and costs. Specifically, the electrolyzer equipment and auxiliary machine parameters include the current prices and floor space occupied by various bidders' electrolyzers and corresponding auxiliary machines. Construction period investment includes wind and solar power investment, hydrogen production investment, energy storage investment, power transmission investment, and various land acquisition fees and preliminary expenses payable during the construction period. Operation period revenue includes hydrogen price, hydrogen volume, electricity (if any), and other potential revenue items such as oxygen sales. Operation period costs include personnel costs, material costs, repair costs, insurance premiums, construction period interest, periodic rental payments during the operation period, energy storage core replacement costs, electrolyzer overhaul costs, depreciation, and safety production costs, other manufacturing costs, other management costs, water and electricity costs, and catalyst costs associated with hydrogen production.

[0199] Preferably, the data preprocessing is performed using the same steps as S1, and the processed data is represented as D2.

[0200] S4: Using the preprocessed wind-solar coupled data, compare the monthly output differences in each year, select typical data for each month from the data of each year, and piece them together.

[0201] Preferably, the specific process of step S4 is as follows:

[0202] S41. Construct a data D3 to organize the wind and solar power output for each month of each year.

[0203] S42. In D3, compare the data at each timestamp and use machine learning to filter the data to obtain typical data for the current month's calculations.

[0204] S43. Combine the filtered monthly wind and solar power output data to form the annual 8760 wind and solar power output curve D4 (e.g., wind and solar power output curve).

[0205] S5: Using the pieced-together data, the power output capacity of wind and solar power is further subdivided according to gradients.

[0206] Preferably, the data used in step S5 is the D4 dataset that was processed in S4. When subdividing the wind and solar power output ratio according to the gradient, the step size S (e.g., gradient step size) needs to be determined first. After determining S, D4 is subdivided according to the step size S to form the base load LB (e.g., base load component) and the fluctuating load LF (e.g., fluctuating load component), and D5 = (S, LB, LF), where LB + LF = D4.

[0207] S6: Based on the principle that large standard cubic meter electrolyzers meet the base load and small standard cubic meter electrolyzers respond to wind and solar fluctuations (for example, based on preset matching rules), calculations are performed according to the step size of the previous step, combined with the data required for the evaluation calculation.

[0208] Preferably, the specific process of step S6 is as follows:

[0209] S61. The basic principle is to use large standard cubic meter electrolytic cells to meet the base load, and small standard cubic meter electrolytic cells to respond to wind and solar fluctuations. That is, LB uses large standard cubic meter electrolytic cells and LF uses small standard cubic meter electrolytic cells.

[0210] S62. Using the electrolytic cell data collected in D2 as input parameters, calculate the data in D5 one by one according to the principle of S61 and the step size S determined in S5 to form multiple configuration schemes.

[0211] S63. Combine the configuration scheme formed in S62 with the parameters in D2 to calculate the economic reference indicators of the project.

[0212] S7: Calculate the optimal electrolytic cell configuration scheme (e.g., the target configuration scheme) based on the principle of optimal economic efficiency.

[0213] Preferably, step S7 uses the IRR formed in S63 as a reference index to select the optimal configuration scheme of different standard electrolyzers.

[0214] The above scenario examples verify the electrolyzer configuration method based on fluctuating wind-solar coupled power generation provided in this manual. The basic principle is to use large standard-sized electrolyzers as the base load and small standard-sized electrolyzers to respond to the fluctuations of new energy sources. By configuring electrolyzers of different standard-sized units in a tiered manner to maximize the utilization of the electrolyzers' own equipment characteristics, the configuration of electrolyzers can be accurate and reasonable, which can effectively increase the rate of return of new energy power plant projects.

[0215] While this specification provides the steps of operation for the methods described in the embodiments or flowcharts, more or fewer steps may be included based on conventional or non-inventive means. The order of steps listed in the embodiments is merely one possible order of execution among many steps and does not represent the only possible order. In actual device or client product execution, the methods shown in the embodiments or drawings may be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment, or even a distributed data processing environment). The terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitations, the presence of other identical or equivalent elements in a process, method, product, or apparatus that includes said elements is not excluded. The terms "first," "second," etc., are used to denote names and do not indicate any particular order.

[0216] Those skilled in the art will also know that, besides implementing the controller using purely computer-readable program code, the same functions can be achieved by logically programming the method steps, making the controller function as logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers (PLCs), and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices within it used to implement various functions can also be considered structures within that hardware component. Alternatively, the devices used to implement various functions can be considered as both software modules implementing the method and structures within a hardware component.

[0217] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer-readable storage media, including storage devices.

[0218] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this specification can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions of this specification can essentially be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments of this specification.

[0219] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. This specification can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0220] Although this specification has been described by way of examples, those skilled in the art will recognize that many variations and modifications are possible without departing from the spirit of this specification, and it is intended that the appended claims cover such variations and modifications without departing from the spirit of this specification.

Claims

1. A method for configuring and processing electrolytic cells based on wave-like wind-solar coupled power generation, characterized in that, include: Obtain historical landscape data of the target area; wherein, the historical landscape data includes landscape data at multiple historical time points, and the landscape data at each historical time point also carries a timestamp corresponding to the historical time point; Identify decision-making impact data for the target region; According to the preset construction rules, the wind and solar power output curve is constructed using historical wind and solar data of the target area; wherein, the wind and solar power output curve is used to characterize the reference power of wind and solar coupled power generation at various time points within a year. Based on the wind and solar power output curve, a matching gradient step size is determined; and using this gradient step size, the wind and solar power output curve is processed to separate the base load component and the fluctuating load component. Based on preset matching rules and gradient step size, optimization is performed according to the base load component, fluctuating load component, and decision influence data to determine the target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolytic cell, and the fluctuating load component is preferentially matched with the second standard square electrolytic cell, wherein the standard square of the first standard square electrolytic cell is greater than that of the second standard square electrolytic cell. According to the target configuration scheme, target electrolytic cells are deployed for the target wind-solar coupled power station to store the electrical energy generated by wind-solar coupled power generation.

2. The method according to claim 1, characterized in that, The decision-influencing data includes at least one of the following: parameters of electrolytic cell equipment and auxiliary equipment, construction costs, operating costs, and operating revenue; Accordingly, determine the decision-making impact data for the target area, including: Obtain electricity price data and power grid rules for the target area; By combining electricity price data and grid rules for the target area, decision-making impact data for the target area can be determined.

3. The method according to claim 1, characterized in that, Based on preset construction rules, and using historical landscape data of the target area, a landscape output curve is constructed, including: Based on the timestamp, multiple power generation data groups are divided using historical landscape data of the target area; each power generation data group corresponds to a month, containing landscape data of multiple historical time points of that month in different years, as well as landscape data of multiple historical time points of different months in the same year that are adjacent to that month. Based on multiple power generation data sets, multiple corresponding power generation data matrices are constructed; wherein each of the multiple power generation data matrices corresponds to a month. Multiple power generation data matrices are processed using a pre-defined typical data generation model to obtain reference data sets for multiple wind and solar power generation in different months. By splicing together and utilizing multiple sets of reference data on wind and solar power generation, the corresponding wind and solar power output curves are obtained.

4. The method according to claim 1, characterized in that, Based on the wind and solar power output curve, determine the matching gradient step size, including: The wind and solar power output curve is subjected to graphic feature extraction to obtain the corresponding graphic features; Based on the aforementioned graphic features, the fluctuation characteristics of the wind and solar power output curve are determined; Based on the fluctuation characteristics, a matching gradient step size is determined.

5. The method according to claim 1, characterized in that, Based on preset matching rules and gradient step size, and according to the base load component, fluctuating load component, and decision impact data, optimization is performed to determine a target configuration scheme that meets the requirements, including: Construct a set of objective functions based on decision impact data; and construct objective constraints based on preset matching rules, base load components, and fluctuating load components. Based on the preset processing rules, multiple computing nodes and the initial solutions for each computing node are determined. Multiple computing nodes are invoked to perform multiple rounds of iterative solution for the objective function set based on the objective constraints and initial solution, so as to obtain the corresponding objective processing result; Based on the target processing results, a target configuration scheme that meets the requirements is determined.

6. The method according to claim 5, characterized in that, Multiple computing nodes are invoked to perform multiple rounds of iterative solutions to the objective function set based on the stated objective constraints and initial solution, including: Multiple computing nodes can be invoked for the current iteration in the following manner: Obtain and filter out the solutions for the current round that meet the retention criteria based on the solutions from the previous round of each computing node; The computing nodes that hold the retained solution for the current round are marked as first-class nodes, and the other computing nodes among the multiple computing nodes, excluding the first-class nodes, are marked as second-class nodes; The retained solutions of the current round are processed by a preset encoding to obtain the corresponding encoded data; and based on the encoded data, the solutions are modified to obtain multiple modified codes. The modification code is decoded to obtain multiple modification solutions for the current round; and the multiple modification solutions for the current round are assigned to the corresponding second-type nodes. The first type of node and the second type of node are called to perform iterative solutions based on the local iterative solution rules and gradient step size, respectively using the retained solution and the modified solution of the current round, to obtain the solution of each computing node in the current round.

7. The method according to claim 6, characterized in that, After calling the first type of nodes and the second type of nodes to iteratively solve the problem according to the local iterative solution rules and gradient step size, using the retained solution and the modified solution of the current round respectively, to obtain the solution of each computing node for the current round, the method further includes: Check whether the termination condition of the iterative solution is met; If the termination condition of the iterative solution is met, the solution with the highest matching degree with the objective function set is selected from the solutions of the current round of multiple computing nodes and used as the objective processing result.

8. An electrolytic cell configuration and processing device based on wave-like wind-solar coupled power generation, characterized in that, include: The acquisition module is used to acquire historical landscape data of the target area; wherein, the historical landscape data includes landscape data at multiple historical time points, and the landscape data at each historical time point also carries a timestamp corresponding to the historical time point; The determination module is used to determine the decision impact data for the target area; The construction module is used to construct a wind and solar power output curve based on the historical wind and solar data of the target area according to the preset construction rules; wherein the wind and solar power output curve is used to characterize the reference power of wind and solar coupled power generation at various time points within a year. The splitting module is used to determine the matching gradient step size based on the wind and solar power output curve; and to use the gradient step size to process the wind and solar power output curve and split the base load component and the fluctuating load component. The solution module is used to perform optimization solutions based on preset matching rules and gradient step size, according to the base load component, fluctuating load component, and decision influence data, to determine the target configuration scheme that meets the requirements; wherein, the preset matching rules include: the base load component is preferentially matched with the first standard square electrolytic cell, and the fluctuating load component is preferentially matched with the second standard square electrolytic cell, wherein the standard square of the first standard square electrolytic cell is greater than that of the second standard square electrolytic cell; The deployment module is used to deploy target electrolytic cells for the target wind-solar coupled power station according to the target configuration scheme, in order to store the electrical energy generated by wind-solar coupled power generation.

9. A server, characterized in that, It includes a processor and a memory for storing processor-executable instructions, wherein the processor, when executing the instructions, implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores computer instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images