A wind-solar coupling hydrogen production and consumption integrated optimization system and method

Abstract

The present application relates to a kind of wind and light coupling hydrogen production and consumption integration optimization system and method, it is related to new energy power generation and hydrogen production technical field, wherein, including: wind and light coupling output simulation module, for based on multi-source wind and light resource data, generation wind and light timing simulation output data;Electrolytic cell optimized hydrogen production module, for according to wind and light timing simulation output data, with minimizing wind and light abandoned power rate and maximizing hydrogen production efficiency as optimization goal, under the condition that wind and light output rate of change, electrolytic cell load following rate, chemical load adjustment rate meet three rate coupling matching constraint, determine electrolytic cell hydrogen production scheme;Flexible chemical consumption module, for constructing the consumption model for representing the mapping relationship between hydrogen supply and chemical product yield under multiple loads;Integrated optimization module, for based on electrolytic cell hydrogen production scheme and consumption model, according to wind and light installed capacity scale or target chemical product yield, optimization system configuration scheme and operation control strategy.

Description

Technical Field

[0001] The embodiments in this specification relate to the fields of new energy power generation and hydrogen production technology, specifically to an integrated optimization system and method for wind-solar coupled hydrogen production and consumption. Background Technology

[0002] Wind-solar coupled hydrogen production is the mainstream path for green hydrogen production, which can not only improve the level of new energy consumption, but also provide clean fuel for industries, transportation and other fields.

[0003] However, the intermittent and fluctuating nature of wind and solar power output in current wind-solar coupled hydrogen production units is inherently contradictory to the demand for stable loads in the hydrogen production process for the consumption unit. It is difficult to achieve integrated optimization of wind-solar coupled hydrogen production and consumption, and the problem of wind and solar curtailment is serious. Summary of the Invention

[0004] The purpose of the embodiments in this specification is to provide an integrated optimization system and method for wind and solar coupled hydrogen production and consumption, so as to overcome the problems of existing methods that make it difficult to achieve integrated optimization of wind and solar coupled hydrogen production and consumption, and the serious problem of wind and solar curtailment.

[0005] To address the aforementioned technical problems, this specification provides, in one aspect, an integrated optimized system for wind-solar coupled hydrogen production and consumption, comprising: The wind-solar coupled output simulation module is used to generate wind-solar time-series simulated output data based on multi-source wind and solar resource data; The electrolyzer optimization hydrogen production module is used to determine the electrolyzer hydrogen production scheme based on the wind and solar power output data simulated over time, with the optimization objectives of minimizing the wind and solar power curtailment rate and maximizing the hydrogen production efficiency. Under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints. The flexible chemical processing module is used to construct a processing model that characterizes the mapping relationship between hydrogen supply and chemical product output under various load conditions. The integrated optimization module is used to optimize system configuration and operation control strategies based on the electrolyzer hydrogen production scheme and consumption model, according to the scale of wind and solar power installations or the target chemical product output.

[0006] Furthermore, the integrated optimization module is used for: If the installed capacity of wind and solar power is given, the time-series power generation data of the wind and solar power generation units are determined based on the given installed capacity and the time-series simulated power output data; based on the time-series power generation data, the first system configuration scheme, the first operation control strategy, and the total power demand of the consumption unit are determined. Based on the total electricity demand and the wind and solar power output data from the time-series simulation, the required wind and solar installed capacity, the second system configuration scheme, and the second operation control strategy are derived. The inverted wind and solar installed capacity is used as the new given wind and solar installed capacity. Based on the new given wind and solar installed capacity, the steps of determining the first system configuration scheme and the first operation control strategy, and determining the second system configuration scheme and the second operation control strategy are iteratively executed until the preset termination condition is met, and the third system configuration scheme and the third operation control strategy are obtained.

[0007] Furthermore, the integrated optimization module is also used for: If the installed capacity of wind and solar power is not given, determine the total power demand of the consumption unit based on the output of the target chemical products; based on the total power demand and the time-series simulated output data of wind and solar power, inversely derive the required installed capacity of wind and solar power, the configuration scheme of the second system, and the second operation control strategy. Based on the inverted wind and solar installed capacity and the wind and solar time-series simulated output data, the time-series power generation data of the wind and solar power generation units are determined; based on the time-series power generation data, the first system configuration scheme, the first operation control strategy, and the total power consumption demand of the consumption unit are determined. The determined total electricity demand is taken as the new total electricity demand; based on the new total electricity demand, the steps of determining the second system configuration scheme and the second operation control strategy, as well as determining the first system configuration scheme and the first operation control strategy, are iteratively executed until the preset termination condition is met, and the third system configuration scheme and the third operation control strategy are obtained.

[0008] Furthermore, the integrated optimization module is also used for: Calculate the initial time-series power generation data based on the installed capacity of wind and solar power and the time-series simulated power output data of wind and solar power. Based on initial time-series power generation data, the charging and discharging strategies of the energy storage unit and the power following process of the hydrogen production unit are simulated. When the power change rate in the initial time-series power generation data exceeds the power tracking rate threshold of the hydrogen production unit during the simulation, the amount of power wasted due to peak shaving and valley filling is determined. Based on the initial time-series power generation data and the amount of power wasted after peak shaving and valley filling, the corrected time-series power generation data is generated.

[0009] Furthermore, the integrated optimization module is also used for: With the optimization objectives of minimizing wind and solar curtailment rates and maximizing hydrogen production efficiency, and with the operational constraints of the hydrogen production unit, the charging and discharging efficiency constraints of the energy storage unit, and the three-rate coupling matching constraints as conditions, a first configuration optimization model is constructed. Based on time-series power generation data, a first configuration optimization model is solved to obtain a first system configuration scheme; the first system configuration scheme includes the first number of electrolyzers in operation of the hydrogen production unit, a first power allocation strategy, and a first hourly hydrogen production rate. The total power demand of the consumption unit is determined based on the first hourly hydrogen production of each electrolyzer.

[0010] Furthermore, the integrated optimization module is also used for: With the optimization objectives of minimizing the installed capacity of wind and solar power, minimizing the curtailment rate of wind and solar power, and maximizing the hydrogen production efficiency, and with the hydrogen consumption constraints of the consumption unit and the three-rate coupling matching constraints as conditions, a second configuration optimization model is constructed; the hydrogen consumption constraints include the hydrogen demand range constraints and the hydrogen demand adjustment rate constraints of the consumption unit. Solve the second configuration optimization model to obtain the minimum wind and solar installed capacity and the second system configuration scheme that meet the total power demand; the second system configuration scheme includes the second number of electrolyzers in operation of the hydrogen production unit, the second power allocation strategy, and the second hourly hydrogen production.

[0011] Furthermore, the wind-solar coupling output simulation module is used for: The multi-source wind and solar resource data are sampled at multiple preset time scales; Waveform clustering was performed on the sampling results at each time scale; The clustering results at different time scales are fused to generate the wind and solar time-series simulated power output data.

[0012] Furthermore, the wind-solar coupling output simulation module is also used for: The first response time constant of the PEM electrolyzer and the second response time constant of the ALK electrolyzer in the hydrogen production unit were obtained. A first sampling time scale is determined based on a first response time constant, and a second sampling time scale is determined based on a second response time constant; the first sampling time scale is smaller than the second sampling time scale. Multi-source wind and solar resource data are sampled at the first sampling time scale and the second sampling time scale respectively to obtain a first sampling sequence for characterizing high-frequency fluctuation components and a second sampling sequence for characterizing low-frequency fluctuation components.

[0013] Furthermore, the wind-solar coupling output simulation module is also used for: Based on the hydrogen demand change rate threshold of the consumption unit, the power fluctuation tolerance threshold of the wind and solar power generation unit is determined. Based on the rate of power change between adjacent sampling times in each time scale downsampled sequence, the time scale downsampled sequence is divided into normal fluctuation sequence and extreme fluctuation sequence according to the power fluctuation tolerance threshold. Waveform clustering was performed on multiple waveform sequences included in the normal fluctuation sequence and the extreme fluctuation sequence to obtain multiple classes of normal waveform sequences and multiple classes of extreme waveform sequences.

[0014] Furthermore, the wind-solar coupling output simulation module is also used for: By splicing together multiple types of conventional waveform sequences at each time scale in chronological order, a conventional time-series power output sequence at that time scale is obtained. Based on the power fluctuation tolerance threshold, amplitude scaling or time stretching is performed on multiple extreme waveform sequences at each time scale to obtain the corrected extreme time series output sequence at that time scale. The conventional time-series power output sequences at various time scales are fused layer by layer according to the time scale order to obtain the basic time-series power output data. The modified extreme time series output sequences at each time scale are superimposed onto the basic time series output data in reverse order according to the time scale to obtain the wind and solar time series simulated output data.

[0015] Furthermore, the wind-solar coupling output simulation module is also used for: Acquire meteorological grid data; the meteorological grid data includes first meteorological grid data and second meteorological grid data; Based on the measured meteorological time-series data corresponding to the first meteorological grid data, the first meteorological grid data is corrected to obtain the third meteorological grid data; the measured meteorological time-series data includes measured wind speed time-series data and measured solar radiation time-series data. For each second meteorological grid, select the fourth meteorological grids that are closest to the second meteorological grid from among the multiple third meteorological grids; The weighting coefficients for each fourth meteorological grid are determined based on the distance between each fourth meteorological grid and the second meteorological grid; the weighting coefficients are negatively correlated with the distance. Based on the meteorological values ​​and weighting coefficients of multiple fourth meteorological grids, the meteorological values ​​of the second meteorological grid are corrected to obtain the data of the fifth meteorological grid; Multi-source wind and solar resource data were determined based on data from the third and fifth meteorological grids.

[0016] Furthermore, the wind-solar coupling output simulation module is also used for: The meteorological deviation at each time point is obtained by comparing the data from the first meteorological grid with the measured meteorological time series data. A time window is constructed based on the meteorological fluctuation index of the region where the first meteorological grid data is located; the length of the time window is positively correlated with the meteorological fluctuation index. The meteorological deviation index is obtained by fitting the meteorological deviation at each time point within the time window. The meteorological deviation index is superimposed on the first meteorological grid data to obtain the third meteorological grid data.

[0017] Furthermore, the electrolyzer-optimized hydrogen production module is used for: Based on the wind and solar power output time-series simulation data, the wind and solar power output change rate sequence, the electrolyzer load following rate sequence of the hydrogen production unit, and the hydrogen demand adjustment rate sequence of the consumption unit are determined. Based on the wind and solar power output time-series simulation data, the wind and solar power output change rate sequence, the wind and solar power generation sequence of the wind and solar power generation unit, the hydrogen production sequence of the hydrogen production unit and the load following rate sequence of the electrolyzer, the hydrogen storage sequence of the hydrogen storage system, the energy storage sequence of the energy storage unit, the hydrogen demand sequence of the consumption unit and the hydrogen demand regulation rate sequence are determined. Calculate the first degree of matching among the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence; Calculate the second degree of matching among the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence; The third matching degree is determined based on the first and second matching degrees; With the optimization objectives of minimizing wind and solar curtailment rates and maximizing hydrogen production efficiency, an electrolyzer hydrogen production scheme is determined under the condition that the third matching degree satisfies the three-rate coupling matching constraint.

[0018] Furthermore, the flexible chemical disposal module is used for: Fit the first conditional distribution when the hydrogen supply is greater than the first hydrogen supply threshold under the first load; the first conditional distribution is used to characterize the non-linear growth of chemical product output with a slowing trend and no upper bound tail characteristic. Based on historical hydrogen consumption rates, a second conditional distribution of hydrogen supply under the second load is determined; the second conditional distribution is used to characterize the baseline characteristics of linear growth in chemical product output. The third conditional distribution is fitted when the hydrogen supply is less than the second hydrogen supply threshold under the third load; the third conditional distribution is used to characterize the tail characteristics of chemical product output, which has an upper bound and a non-linear, steep decline. Based on the first conditional distribution, the second conditional distribution, and the third conditional distribution, an absorption model is constructed. Among them, the first load is greater than the second load, and the second load is greater than the third load.

[0019] Furthermore, the embodiments of this specification provide an integrated optimization method for wind-solar coupled hydrogen production and consumption, including: Based on multi-source wind and solar resource data, wind and solar time-series simulated power output data is generated; Based on the simulated output data of wind and solar power, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, the electrolyzer hydrogen production scheme is determined under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints. Construct a consumption model to characterize the mapping relationship between hydrogen supply and chemical product output under various load conditions; Based on the electrolyzer hydrogen production scheme and consumption model, the system configuration scheme and operation control strategy are optimized according to the wind and solar power installed capacity or the target chemical product output.

[0020] Furthermore, embodiments of this specification provide a computer device, including: Memory, used to store computer programs; A processor is used to execute the computer program to implement the above-mentioned integrated optimization method for wind-solar coupled hydrogen production and consumption.

[0021] Furthermore, embodiments of this specification provide a computer storage medium storing computer program instructions, which, when executed, implement the aforementioned integrated optimization method for wind-solar coupled hydrogen production and consumption.

[0022] In another aspect, embodiments of this specification provide a computer program product, including a computer program that, when executed by a processor, implements the aforementioned integrated optimization method for wind-solar coupled hydrogen production and consumption.

[0023] As can be seen from the technical solutions provided in the embodiments of this specification above, the embodiments of this specification can generate wind and solar time-series simulated power output data based on multi-source wind and solar resource data; based on the wind and solar time-series simulated power output data, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, and under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints, an electrolyzer hydrogen production scheme is determined; a consumption model is constructed to characterize the mapping relationship between hydrogen supply and chemical product output under multiple loads; based on the electrolyzer hydrogen production scheme and consumption model, the system configuration scheme and operation control strategy are optimized according to the wind and solar installed capacity or the target chemical product output. By using a three-rate coupling matching constraint, the power output of wind and solar power, the load of electrolyzers, and the chemical industry consumption are dynamically coordinated to avoid power curtailment or frequent equipment start-ups and shutdowns caused by fluctuations and mismatches, thereby improving the stability of the power generation and hydrogen production sides. Furthermore, based on the mapping relationship between hydrogen supply and product output under various load conditions, the adaptability of the chemical industry side to fluctuating hydrogen sources is enhanced. The integrated two-way optimization flexibly configures the system according to the installed capacity of wind and solar power or the target output, realizing the integrated coordinated optimization of power generation, hydrogen production, and consumption. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below.

[0025] Figure 1 This is a schematic diagram of the structural composition of an integrated optimized system for wind and solar coupled hydrogen production and consumption, provided in the embodiments of this specification. Figure 2This is a flowchart of an integrated optimization method for wind-solar coupled hydrogen production and consumption provided in the embodiments of this specification; Figure 3 This is a schematic diagram of the structural composition of a computer device provided in the embodiments of this specification. Detailed Implementation

[0026] 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.

[0027] It should be noted that the terms "first," "second," etc., used in this specification, claims, and the foregoing drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0028] In some embodiments, the above-described integrated wind-solar hydrogen production and consumption optimization system can be applied to an integrated wind-solar hydrogen production and consumption system. The integrated wind-solar hydrogen production and consumption system may include a wind-solar power generation unit, a hydrogen production unit, a consumption unit, a hydrogen storage unit, and an energy storage unit; the wind-solar power generation unit may include wind turbine generators and photovoltaic arrays; the hydrogen production unit may include PEM electrolyzers and ALK electrolyzers; and the consumption unit may include an ammonia synthesis unit, a methanol synthesis unit, or a green aviation fuel synthesis unit.

[0029] The wind and solar power generation unit consists of wind turbine generators and photovoltaic arrays. The wind turbine generators convert wind energy into alternating current (AC) electricity, and the photovoltaic arrays convert solar energy into direct current (DC) electricity. The hydrogen production unit consists of a proton exchange membrane (PEM) electrolyzer and an alkaline (ALK) electrolyzer. The PEM electrolyzer is used to rapidly track power fluctuations and produce hydrogen, while the alkaline electrolyzer is used for low-cost, large-scale hydrogen production under steady-state conditions. The consumption unit consists of at least one of ammonia synthesis, methanol synthesis, or green aviation fuel synthesis. The ammonia synthesis unit reacts hydrogen and nitrogen to produce ammonia; the methanol synthesis unit reacts hydrogen and carbon dioxide to produce methanol; and the green aviation fuel synthesis unit reacts hydrogen and biomass syngas to produce aviation fuel. The hydrogen storage unit consists of one or more hydrogen storage tanks to store hydrogen produced by the hydrogen production unit and release it when the consumption unit needs hydrogen. The energy storage unit consists of an electrochemical energy storage battery array to smooth out instantaneous power fluctuations in the wind and solar power generation unit and provide system startup power.

[0030] In some embodiments, the system configuration scheme may include a wind and solar power generation scheme, a hydrogen production scheme, a hydrogen storage scheme, an energy storage scheme, and a consumption scheme. The wind and solar power generation configuration scheme may include wind power capacity and photovoltaic capacity. The hydrogen production configuration scheme may include the number of PEM electrolyzers and ALK electrolyzers, their rated power per unit, and the power ratio of the mixture. The hydrogen storage configuration scheme may include the volume of the hydrogen storage tank and its rated hydrogen charging / discharging rate. The energy storage configuration scheme may include the rated power and rated capacity of the electrochemical energy storage. The consumption configuration scheme may include the design capacity, load regulation range, and hydrogen consumption rate of the downstream chemical plant. The operation control strategy may include the hourly start / stop and power allocation strategy for the electrolyzers, the hourly charging / discharging strategy for the hydrogen storage tanks, the hourly charging / discharging strategy for the electrochemical energy storage, and the hourly load regulation strategy for the chemical plant.

[0031] The wind and solar power generation configuration includes two parameters: wind power installed capacity and photovoltaic installed capacity. Wind power installed capacity represents the total rated power of the wind turbine generators, and photovoltaic installed capacity represents the total rated power of the photovoltaic array. The hydrogen production configuration includes the number of proton exchange membrane electrolyzers and alkaline electrolyzers, their rated power per unit, and the mixed power ratio. The mixed power ratio is defined as the ratio of the total rated power of the proton exchange membrane electrolyzers to the total rated power of the alkaline electrolyzers. The hydrogen storage configuration includes the volume of the hydrogen storage tank and its rated charge / discharge rate. The volume represents the total amount of hydrogen that can be stored, and the rated charge / discharge rate represents the maximum amount of hydrogen that can be charged or released per unit time. The energy storage configuration includes the rated power and rated capacity of the electrochemical energy storage. The rated power represents the maximum instantaneous output electrical energy of the energy storage system, and the rated capacity represents the total electrical energy that the energy storage system can store. The consumption configuration plan includes the design capacity, load adjustment range, and hydrogen consumption rate of the downstream chemical plant. Design capacity represents the product output per unit time under rated operating conditions; load adjustment range represents the minimum and maximum load intervals for safe operation; and hydrogen consumption rate represents the amount of hydrogen consumed per unit of product produced. Operational control strategies include hourly start-up and shutdown and power allocation strategies for electrolyzers, hourly charging and discharging strategies for hydrogen storage tanks, hourly charging and discharging strategies for electrochemical energy storage, and hourly load adjustment strategies for the chemical plant.

[0032] This specification provides an integrated optimization system 100 for wind and solar coupled hydrogen production and consumption, including: a wind and solar coupled output simulation module 101, an electrolyzer optimized hydrogen production module 102, a flexible chemical consumption module 103, and an integrated optimization module 104.

[0033] In some embodiments, the wind-solar coupled power output simulation module 101 described above can be used to generate wind-solar time-series simulated power output data based on multi-source wind and solar resource data.

[0034] In some embodiments, the above-mentioned generation of wind and solar time-series simulated power output data based on multi-source wind and solar resource data may specifically include, that is, the wind and solar coupled power output simulation module 101 may be used to: acquire multi-source wind and solar resource data of the target area; sample the multi-source wind and solar resource data at multiple preset time scales; perform waveform clustering on the sampling results at each time scale; and fuse the clustering results at each time scale to generate the wind and solar time-series simulated power output data.

[0035] The wind-solar coupled power output simulation module 101 is used to convert multi-source wind and solar resource data into wind and solar time-series simulated power output data. The multi-source wind and solar resource data includes historical wind speed and solar radiation intensity data from meteorological reanalysis fields, satellite inversion products, and ground observation stations. This module can acquire the aforementioned multi-source data within the target area and sample it at hourly, daily, and seasonal scales. Hourly sampling captures intraday fluctuation characteristics, daily sampling reflects the impact of weather processes, and seasonal sampling characterizes annual periodic patterns. Dynamic time curvature distance clustering is performed on the sampling results at each time scale, grouping time periods with similar waveforms into one category, with each category corresponding to a typical power output pattern. The clustering results at the three time scales are fused using a weighted voting mechanism; that is, for the same moment, the power output value represented by the majority of clustering results is taken as the final value, thereby generating a wind and solar time-series simulated power output data that simultaneously reflects short-term fluctuations, weather processes, and seasonal changes. This approach, by separating and then fusing features at different time scales, avoids the loss of details or amplification of noise at a single scale, improves the representativeness of the output data, and provides a more accurate input boundary for subsequent hydrogen production optimization.

[0036] In some embodiments, the above-mentioned electrolyzer optimization hydrogen production module 102 can be used to determine the electrolyzer hydrogen production scheme based on the wind and solar power output data simulated over time, with the optimization objectives of minimizing wind and solar power curtailment rate and maximizing hydrogen production efficiency, under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraint.

[0037] In some embodiments, the above-mentioned method of determining an electrolyzer hydrogen production scheme based on wind and solar power output time-series simulation data, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate satisfy the three-rate coupling matching constraint, may further include, specifically, the above-mentioned electrolyzer optimized hydrogen production module 102 may be used to: determine the wind and solar power output change rate sequence, the electrolyzer load following rate sequence of the hydrogen production unit, and the hydrogen demand adjustment rate sequence of the consumption unit based on the wind and solar power output time-series simulation data; determine the wind and solar power output change rate sequence, the wind and solar power generation sequence of the wind and solar power generation unit, and the wind and solar power generation sequence based on the wind and solar power output time-series simulation data. The hydrogen production sequence and electrolyzer load following rate sequence of the hydrogen production unit, the hydrogen storage sequence of the hydrogen storage system, the energy storage sequence of the energy storage unit, and the hydrogen demand sequence and hydrogen demand regulation rate sequence of the consumption unit are calculated. The first degree of matching between the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence is calculated. The second degree of matching between the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand regulation rate sequence is calculated. Based on the first and second degree of matching, the third degree of matching is determined. With the optimization objectives of minimizing the wind and solar curtailment rate and maximizing the hydrogen production efficiency, the electrolyzer hydrogen production scheme is determined under the condition that the third degree of matching satisfies the three-rate coupling matching constraint.

[0038] The electrolyzer optimization hydrogen production module 102 can determine the electrolyzer hydrogen production scheme based on wind and solar time-series simulated output data. This module can extract three key sequences from the simulated output data: the wind and solar output change rate sequence (the ratio of the difference in wind and solar power generation at adjacent times to the power at the previous time); the electrolyzer load following rate sequence (the ratio of the actual change in electrolyzer power to the change in wind and solar power); and the downstream chemical plant hydrogen demand adjustment rate sequence (the ratio of the hourly change in hydrogen demand of downstream chemical plants to the rated hydrogen demand). Simultaneously, the module also extracts wind and solar power generation sequences, hydrogen production sequences, hydrogen storage sequences, energy storage sequences, and hydrogen demand sequences. The Pearson correlation coefficient between the first five sequences is calculated as the first degree of matching, used to measure the energy balance between source, storage, and load. The cosine similarity between the wind and solar output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence is calculated as the second degree of matching, used to measure the consistency of their changing directions. The third degree of matching is obtained by multiplying the first and second degrees of matching and taking the square root. When the third matching degree is greater than a preset threshold, the three-rate coupling matching constraint is deemed satisfied. Based on this, with the dual objectives of minimizing the curtailment rate and maximizing hydrogen production efficiency, a non-dominated sorting genetic algorithm is invoked to solve the start-up and shutdown and power allocation schemes of the electrolyzer. This calculation method separately evaluates and then integrates the static energy balance and dynamic change consistency, avoiding the omission of transient mismatches by a single indicator and improving the hydrogen production scheme's ability to adapt to wind and solar power fluctuations.

[0039] In some embodiments, the flexible chemical processing module 103 described above can be used to construct a processing model that characterizes the mapping relationship between hydrogen supply and chemical product output under various load conditions.

[0040] In some embodiments, the above-described absorption model for characterizing the mapping relationship between hydrogen supply and chemical product output under multiple loads may specifically include, that is, the flexible chemical absorption module 103 may be used to: fit a first conditional distribution when the hydrogen supply under a first load is greater than a first hydrogen supply threshold; the first conditional distribution characterizes the tail characteristic of chemical product output showing nonlinear growth with a slowdown and no upper bound; determine a second conditional distribution of hydrogen supply under a second load based on historical hydrogen consumption rates; the second conditional distribution characterizes the baseline characteristic of linear growth in chemical product output; fit a third conditional distribution when the hydrogen supply under a third load is less than a second hydrogen supply threshold; the third conditional distribution characterizes the tail characteristic of chemical product output showing an upper bound and a nonlinear decline with a steepening; and construct an absorption model based on the first conditional distribution, the second conditional distribution, and the third conditional distribution; wherein, the first load is greater than the second load, and the second load is greater than the third load.

[0041] The flexible chemical processing module 103 can construct a mapping model between hydrogen supply and chemical product output. This module divides the chemical load into high-load, medium-load, and low-load zones. In the high-load zone, a conditional distribution is fitted when the hydrogen supply exceeds a first hydrogen supply threshold. A generalized Pareto distribution with a shape parameter greater than zero is used to characterize the tail characteristic where output growth slows down with increasing hydrogen supply and there is no theoretical upper limit to output. In the medium-load zone, a linear function between hydrogen supply and output is established based on the median of historical hydrogen consumption rate data as the baseline hydrogen consumption rate. This characterizes the baseline characteristic where output increases proportionally with hydrogen supply under steady-state operation. In the low-load zone, a conditional distribution is fitted when the hydrogen supply is less than a second hydrogen supply threshold. A generalized Pareto distribution with a shape parameter less than zero is used to characterize the tail characteristic where output has a technical upper limit and output declines rapidly when hydrogen supply decreases. The generalized Pareto distribution function of the high-load region, the linear function of the medium-load region, and the generalized Pareto distribution function of the low-load region are connected by the boundary values ​​of the load intervals and smoothed through cubic spline interpolation to form a piecewise continuous absorption model across the entire domain. This modeling method solves the technical problem that traditional single linear models cannot accurately describe the nonlinear response under extreme high and low loads by distinguishing the differences in tail behavior of different load intervals, thereby improving the downstream absorption capacity to predict and adapt to fluctuating hydrogen sources.

[0042] In some embodiments, the integrated optimization module 104 described above can be used to optimize the system configuration scheme and operation control strategy based on the electrolyzer hydrogen production scheme and consumption model, according to the wind and solar power installed capacity or the target chemical product output.

[0043] In some embodiments, the above-mentioned optimization of system configuration scheme and operation control strategy based on electrolyzer hydrogen production scheme and consumption model, according to wind and solar installed capacity or target chemical product output, may specifically include, that is, the above-mentioned integrated optimization module 104 may be used to: if the wind and solar installed capacity is given, determine the time-series power generation data of the wind and solar power generation unit according to the given wind and solar installed capacity and wind and solar time-series simulated output data; determine the first system configuration scheme, the first operation control strategy and the total power demand of the consumption unit according to the time-series power generation data; invert the required wind and solar installed capacity, the second system configuration scheme and the second operation control strategy according to the total power demand and wind and solar time-series simulated output data; take the inverted wind and solar installed capacity as the newly given wind and solar installed capacity; based on the newly given wind and solar installed capacity, iteratively execute the above steps of determining the first system configuration scheme and the first operation control strategy and determining the second system configuration scheme and the second operation control strategy until the preset termination condition is met, and obtain the third system configuration scheme and the third operation control strategy.

[0044] The integrated optimization module 104 performs forward-backward bidirectional iterative optimization given a fixed wind and solar installed capacity. The module obtains the given wind and solar installed capacity values, combines them with time-series simulation output data, and calculates the total wind and solar power generation hourly to obtain time-series power generation data. Based on this data, it calls the electrolyzer optimization hydrogen production module and the flexible chemical consumption module to determine an initial configuration scheme, including the number and power ratio of electrolyzers, energy storage capacity, and hydrogen storage capacity, and calculates the total electricity demand of the consumption unit. Using the calculated total electricity demand as input, it calls the wind-solar coupling output simulation module to inversely calculate the minimum wind and solar installed capacity required to meet this demand, obtaining a new set of wind and solar capacity values. The new installed capacity obtained from the inversion replaces the original given values, and the steps of forward determination of the initial configuration scheme and backward inversion of the installed capacity are repeated. This iterative process continues until the relative change rate of the total electricity demand output between two adjacent iterations is less than a preset convergence threshold, at which point the current configuration scheme is output as the final system configuration scheme and the corresponding operation control strategy. This bidirectional iterative mechanism resolves the mismatch between the given installed capacity and power demand in traditional unidirectional optimization. By adjusting the source and load, the configuration scheme simultaneously meets both the installed capacity constraint and the power consumption requirement, achieving global optimization.

[0045] In some embodiments, the above-mentioned optimization of system configuration scheme and operation control strategy based on electrolyzer hydrogen production scheme and consumption model, according to wind and solar installed capacity or target chemical product output, may specifically include, that is, the above-mentioned integrated optimization module 104 may also be used for: if the wind and solar installed capacity is not given, determining the total power demand of the consumption unit according to the target chemical product output; based on the total power demand and wind and solar time-series simulated output data, inverting the required wind and solar installed capacity, second system configuration scheme and second operation control strategy; based on the inverted wind and solar installed capacity and wind and solar time-series simulated output data, determining the time-series power generation data of the wind and solar power generation unit; based on the time-series power generation data, determining the first system configuration scheme, first operation control strategy and total power demand of the consumption unit; using the determined total power demand as the new total power demand; based on the new total power demand, iteratively executing the above steps of determining the second system configuration scheme and second operation control strategy and determining the first system configuration scheme and first operation control strategy, until the preset termination condition is met, to obtain the third system configuration scheme and third operation control strategy.

[0046] The integrated optimization module 104 can perform bidirectional iterative optimization (backward and forward) when the wind and solar installed capacity is not given but the output of the target chemical product is known. The module obtains the output of the target chemical product, such as 100,000 tons of green ammonia or 150,000 tons of green methanol per year. It then calls the mapping model in the flexible chemical consumption module to calculate the total annual hydrogen supply required to meet this output. Combined with the unit power consumption of the electrolyzer, it calculates the total power demand of the consumption unit. Using the total power demand as the target, it calls the wind-solar coupling output simulation module to inversely derive the wind and solar installed capacity that meets this power demand and has the lowest life-cycle cost under given regional wind and solar resource conditions, obtaining a set of wind and solar capacity values. Using the inverted installed capacity as a known quantity, it performs forward optimization: combining wind and solar time-series simulation output data, it determines the time-series power generation data, and then calls the electrolyzer optimization hydrogen production module and the flexible chemical consumption module to determine a configuration scheme and calculate the actual total power demand of the consumption unit. The actual electricity demand is compared with the target electricity demand. If the relative deviation exceeds a set tolerance, the actual electricity demand is taken as the new target electricity demand, and the reverse calculation of installed capacity and the forward determination of the configuration scheme are repeated. This process is iterated until the deviation between the actual and target electricity demands converges, and the final configuration scheme and operating strategy are output. This reverse-forward iterative approach, which derives the source-end scale from the end-product demand and then corrects it through forward verification, solves the inefficiency problem of repeated manual calculations required in the traditional sales-driven production model, and achieves automated and accurate solutions from product output to installed capacity configuration.

[0047] In some embodiments, the sampling of the multi-source wind and solar resource data at multiple preset time scales may further include, specifically, the wind-solar coupled output simulation module 101 may be used to: obtain the first response time constant of the PEM electrolyzer and the second response time constant of the ALK electrolyzer in the hydrogen production unit; determine the first sampling time scale based on the first response time constant, and determine the second sampling time scale based on the second response time constant; the first sampling time scale is smaller than the second sampling time scale; and sample the multi-source wind and solar resource data at the first sampling time scale and the second sampling time scale respectively to obtain a first sampling sequence for characterizing high-frequency fluctuation components and a second sampling sequence for characterizing low-frequency fluctuation components.

[0048] The wind-solar coupling output simulation module 101 can acquire the response time constants of the proton exchange membrane electrolyzer and the alkaline electrolyzer in the hydrogen production unit. The response time constant characterizes the time required for the electrolyzer to reach the target output power (e.g., 63.2%) after receiving a power change command. The constant range for the proton exchange membrane electrolyzer is 1 to 60 seconds, and for the alkaline electrolyzer, it is 30 to 300 seconds. The module can multiply the response time constant of the proton exchange membrane electrolyzer by a preset coefficient to obtain an hourly sampling interval, used to capture high-frequency fluctuations from seconds to minutes; and multiply the response time constant of the alkaline electrolyzer by the same coefficient to obtain a daily sampling interval, used to capture low-frequency fluctuations from minutes to hours. The hourly sampling interval is numerically smaller than the daily sampling interval. The module extracts wind speed sequences and solar radiation intensity sequences from multi-source wind and solar resource data at equal intervals according to the hourly and daily sampling intervals, respectively, generating a first sampling sequence and a second sampling sequence. The first sampling sequence preserves the rapid random fluctuations in wind and solar power, while the second sampling sequence preserves the slow changing trends dominated by weather processes. This differentiated sampling method ensures that the generated wind and solar data matches the dynamic response capabilities of different electrolyzers in terms of time granularity, avoiding the inclusion of invalid information that electrolyzers cannot follow in high-frequency data, while retaining low-frequency data for long-term energy storage planning.

[0049] In some embodiments, the above-mentioned waveform clustering of the sampling results at each time scale may further include, specifically, the wind-solar coupled output simulation module 101 may be used to: determine the power fluctuation tolerance threshold of the wind-solar power generation unit in the wind-solar-hydrogen storage system based on the hydrogen demand change rate threshold of the absorption unit; divide the sampling sequence at each time scale into a normal fluctuation sequence and an extreme fluctuation sequence based on the power change rate between adjacent sampling times in the sampling sequence at each time scale and the power fluctuation tolerance threshold; and perform waveform clustering on the multiple waveform sequences included in the normal fluctuation sequence and the extreme fluctuation sequence to obtain multiple types of normal waveform sequences and multiple types of extreme waveform sequences.

[0050] The wind-solar coupling output simulation module 101 can determine the power fluctuation tolerance threshold of the wind and solar power generation unit based on the hydrogen demand change rate threshold of the absorption unit. The hydrogen demand change rate threshold can be given by the design parameters of the downstream chemical plant, for example, the allowable load change of the ammonia synthesis unit per hour does not exceed 20% of the rated load. The module multiplies this percentage threshold by the rated total power of the wind and solar power generation unit to obtain an absolute value boundary of the power change rate, which serves as the power fluctuation tolerance threshold. For each time scale sampling sequence (e.g., hourly scale sequence), the module iterates through two adjacent sampling points in the sequence and calculates the absolute value of the power change rate. Waveform segments between adjacent points with a change rate less than or equal to the threshold are marked as normal fluctuation sequences, and waveform segments with a change rate greater than the threshold are marked as extreme fluctuation sequences. The normal fluctuation sequence corresponds to the power fluctuation range that the electrolyzer and downstream units can normally follow, while the extreme fluctuation sequence corresponds to severe fluctuations that exceed the system's dynamic adjustment capability. The module performs subsequent clustering processing on multiple waveform segments extracted from the normal fluctuation sequence and the extreme fluctuation sequence, respectively. This classification method directly links the severity of wind and solar power fluctuations to the actual capacity of downstream chemical plants, enabling different handling strategies to be adopted for different fluctuation types. This avoids the waste of new energy or over-investment caused by using a uniform conservative approach to all fluctuations.

[0051] In some embodiments, the above-mentioned waveform clustering of multiple waveform sequences included in the conventional fluctuation sequence and the extreme fluctuation sequence to obtain multiple classes of conventional waveform sequences and multiple classes of extreme waveform sequences may further include, specifically, the wind-solar coupling output simulation module 101 may be used to: obtain the power cycle life curve of the electrolyzer in the hydrogen production unit, wherein the power cycle life curve is used to characterize the cumulative damage threshold of the electrolyzer under different power change amplitudes and cycle numbers; for each waveform sequence in the conventional fluctuation sequence and the extreme fluctuation sequence, extract the power change amplitude, change duration and average power of the waveform sequence. The rate of change of average power is calculated, and the equivalent fatigue damage index caused by the waveform sequence to the electrolytic cell is calculated based on the power cycle life curve. The equivalent fatigue damage index is used as the first clustering feature, the average power change rate is used as the second clustering feature, and the duration of change is used as the third clustering feature. Based on the first, second, and third clustering features, multiple waveform sequences in the normal fluctuation sequence and the extreme fluctuation sequence are clustered to obtain multiple normal waveform clusters including low-damage normal waveform clusters and high-damage normal waveform clusters, as well as multiple extreme waveform clusters including short-time extreme waveform clusters and long-time extreme waveform clusters.

[0052] The wind-solar coupling output simulation module 101 can obtain the power cycle life curve of the electrolyzer in the hydrogen production unit. This curve is obtained through accelerated aging tests. The horizontal axis represents the power change amplitude (expressed as a percentage of rated power), and the vertical axis represents the number of cycles allowed at that amplitude. The curve exhibits an exponentially decreasing shape: the larger the change amplitude, the fewer the allowed number of cycles. The module extracts three features for each waveform segment in the normal fluctuation sequence and extreme fluctuation sequence: power change amplitude (the difference between the maximum and minimum power within the segment divided by the rated power), change duration (the time difference between the start and end of the segment), and average power change rate (amplitude divided by duration). For each waveform segment, the module finds the corresponding number of allowed cycles on the life curve based on its power change amplitude, and then calculates the equivalent fatigue damage index based on the expected frequency of the waveform segment in historical data. The calculation formula is the expected occurrence frequency divided by the allowed number of cycles, and the damage indices of multiple amplitude segments are summed. The module uses the equivalent fatigue damage index as the first clustering feature, the average power change rate as the second clustering feature, and the change duration as the third clustering feature. The K-means algorithm was used to cluster waveform segments in both regular and extreme fluctuation sequences. For regular fluctuation sequences, two clusters were obtained: a low-damage regular waveform cluster (damage index less than 0.1, gradual rate of change, and moderate duration) and a high-damage regular waveform cluster (damage index greater than 0.3, steep rate of change, or long duration). For extreme fluctuation sequences, two clusters were obtained: a short-time extreme waveform cluster (large amplitude but duration less than 10 seconds) and a long-time extreme waveform cluster (large amplitude and duration exceeding 60 seconds). This clustering method uses the fatigue life of the electrolyzer as the basis for waveform classification, enabling the subsequently generated wind and solar data to distinguish fluctuation patterns with different levels of damage. This facilitates the intentional avoidance of high-damage waveforms in optimized configurations to extend equipment life.

[0053] In some embodiments, the above-mentioned fusion of clustering results at various time scales to generate the wind and solar time-series simulated power output data may further include, specifically, the wind and solar coupled power output simulation module 101 may be used to: splice multiple types of conventional waveform sequences at each time scale in chronological order to obtain a conventional time-series power output sequence for that time scale; perform amplitude scaling or time stretching processing on multiple types of extreme waveform sequences at each time scale based on the power fluctuation tolerance threshold to obtain a modified extreme time-series power output sequence for that time scale; fuse the conventional time-series power output sequences at each time scale layer by layer in chronological order to obtain basic time-series power output data; and superimpose the modified extreme time-series power output sequences at each time scale in reverse chronological order onto the basic time-series power output data to obtain wind and solar time-series simulated power output data.

[0054] The wind-solar coupled power output simulation module 101 can reconstruct the complete conventional time-series power output sequence for each time scale by splicing together waveform segments from multiple conventional waveform clusters obtained at each time scale according to their original time order. For the extreme waveform sequence at each time scale, the module processes it based on the power fluctuation tolerance threshold: for waveform segments in short-time extreme waveform clusters, amplitude compression is used to proportionally reduce the power amplitude of the waveform segment, so that the maximum power change rate after compression is reduced to below the tolerance threshold, while keeping the time axis of the waveform unchanged; for waveform segments in long-time extreme waveform clusters, time stretching is used to proportionally extend the time axis of the waveform segment, so that the equivalent change rate after stretching is reduced to below the tolerance threshold, while keeping the absolute amplitude of power change unchanged. All processed extreme waveform segments are connected in chronological order to generate the corrected extreme time-series power output sequence for that time scale. The module performs layer-by-layer fusion of conventional sequences in descending order of time scale (e.g., seasonal scale → daily scale → hourly scale): using the seasonal conventional sequence as a baseline, the daily conventional sequence is low-pass filtered to extract the daily periodic component, calibrated, and then superimposed on the seasonal baseline to obtain the daily-seasonal fused sequence; then, the high-frequency components of the hourly conventional sequence are superimposed to obtain the basic time-series power output data. Subsequently, the module superimposes the corrected extreme sequences of each scale onto the basic data in reverse order of ascending time scale (hourly scale → daily scale → seasonal scale). Before each superposition, each extreme event in the extreme sequence is aligned and superimposed according to its time position, based on the current basic data. The final generated wind and solar time-series simulated power output data retains the statistical distribution characteristics of the original wind and solar resources while correcting extreme fluctuations exceeding the equipment's tolerance range, ensuring the physical feasibility of the input data for subsequent hydrogen production scheme optimization. This fusion method balances the realism of wind and solar resources with engineering feasibility, avoiding the failure of optimization schemes due to untrackable extreme fluctuations in the input data.

[0055] In some embodiments, the method for determining the multi-source wind and solar resource data may specifically include, that is, the wind-solar coupling output simulation module 101 may be used to: acquire meteorological grid data; the meteorological grid data includes first meteorological grid data and second meteorological grid data; based on the measured meteorological time-series data corresponding to the first meteorological grid data, correct the first meteorological grid data to obtain third meteorological grid data; the measured meteorological time-series data includes measured wind speed time-series data and measured solar radiation time-series data; for each second meteorological grid, select multiple fourth meteorological grids that are closest to the second meteorological grid from multiple third meteorological grids; determine the weight coefficient corresponding to each fourth meteorological grid according to the distance between each fourth meteorological grid and the second meteorological grid; the weight coefficient is negatively correlated with the distance; correct the meteorological values ​​of the second meteorological grid according to the meteorological values ​​of multiple fourth meteorological grids and the weight coefficient to obtain fifth meteorological grid data; determine the multi-source wind and solar resource data according to the third meteorological grid data and the fifth meteorological grid data.

[0056] The wind-solar coupling output simulation module 101 can determine multi-source wind and solar resource data. The module acquires meteorological grid data covering the target area. This data divides the area into regularly arranged spatial grid cells, with each grid point storing hourly wind speed and solar radiation intensity values. The meteorological grid data includes two types: the first meteorological grid data corresponds to grid locations where meteorological stations or anemometers are actually deployed within the area, and the second meteorological grid data corresponds to grid locations without actual measuring equipment. The module acquires measured wind speed and solar radiation time-series data at the locations of the first meteorological grid data. The measured data is obtained directly from the anemometers and radiometers at the meteorological stations, and its accuracy is higher than that of the grid data. Using the measured data as a benchmark, the module performs subtraction correction on the hourly values ​​in the first meteorological grid data: it calculates the difference between the grid value and the measured value at each time point, subtracts this difference as a system bias from the original grid values, and obtains the corrected third meteorological grid data. For each second meteorological grid, the module calculates the planar Euclidean distance between that grid and all third meteorological grids, and selects the four third meteorological grids with the smallest distances as the fourth meteorological grid. The module calculates a weighting coefficient based on the distance from each fourth meteorological grid to its corresponding second meteorological grid. This weighting coefficient is equal to the reciprocal of the square of that distance divided by the sum of the reciprocals of the squares of the four distances, thus increasing the weight of grids that are closer together. The module then multiplies the wind speed (or solar radiation) value of each fourth meteorological grid by its weighting coefficient and sums the results to obtain the corrected value for that second meteorological grid. All second meteorological grids are processed in this way to form the fifth meteorological grid data. The module merges the third and fifth meteorological grid data to obtain multi-source wind and solar resource data covering the entire region and corrected using measured data. This correction method uses high-precision measured data to correct wide-area grid data, and then transfers the correction effect to areas without measured data through inverse distance weighting. This solves the technical problem of large local deviations in meteorological reanalysis data under complex terrain, making it unsuitable for direct use in project micro-site selection. It improves the spatial resolution and accuracy of wind and solar resource data during the planning stage.

[0057] In some embodiments, the above-mentioned amplitude scaling or time stretching processing of multiple extreme waveform sequences at each time scale based on the power fluctuation tolerance threshold to obtain the corrected extreme time-series power output sequence for that time scale may further include, specifically, the wind-solar coupled power output simulation module 101 may be used to: obtain the cumulative operating time and health status index of the electrolyzer in the hydrogen production unit; update the power fluctuation tolerance threshold according to the cumulative operating time and health status index; the health status index decays as the power cycle number of the electrolyzer increases; and calculate the power of each waveform segment in each extreme waveform sequence. The ratio of the rate of change to the updated power fluctuation tolerance threshold is used. If the ratio is greater than the first preset threshold, the waveform segment is amplitude compressed, and the compression ratio is set to the reciprocal of the ratio multiplied by a safety factor, so that the compressed power change rate is reduced to below the updated power fluctuation tolerance threshold. If the ratio is less than or equal to the first preset threshold but greater than the second preset threshold, the waveform segment is time-stretched, and the stretching factor is set to the ratio, so that the stretched power change rate is reduced to below the updated power fluctuation tolerance threshold. All processed waveform segments are connected in chronological order to generate a corrected extreme time-series output sequence for that time scale.

[0058] The wind-solar coupling output simulation module 101 can dynamically adjust the electrolyzer's aging state by incorporating amplitude scaling or time stretching when performing amplitude scaling or time stretching on extreme waveform sequences. The module acquires the cumulative operating time (in hours) and health status index of the electrolyzer in the hydrogen production unit. The initial value of the health status index is 1, indicating a brand-new state, and it decays thereafter as the number of power cycles experienced by the electrolyzer increases. A power cycle can be a start-up / shutdown or load increase / decrease process where the power change exceeds a preset proportion of the rated power. The decay pattern is determined by the power cycle life curve provided by the electrolyzer manufacturer: after each effective cycle, the health status index is multiplied by a decay factor less than 1 (e.g., 0.9999 to 0.99999, corresponding to thousands of cycles allowed within the design life). The module multiplies the original power fluctuation tolerance threshold (converted from the downstream hydrogen demand change rate threshold) by the current health status index to obtain the updated power fluctuation tolerance threshold. As the electrolyzer ages, the tolerance threshold decreases, meaning that the aging equipment's ability to withstand power fluctuations decreases. For each waveform segment (a continuous over-threshold fluctuation trajectory) in the extreme waveform sequence, the module calculates the ratio of the maximum power change rate of that segment to the updated threshold. If this ratio is greater than a first preset threshold (e.g., 1.5), the module performs amplitude compression on the segment: the compression ratio is set to the reciprocal of the ratio multiplied by a safety factor, so that the power change rate at each point within the compressed segment is reduced to below the updated threshold, while keeping the time axis of the waveform unchanged. If the ratio is less than or equal to the first preset threshold but greater than a second preset threshold (e.g., 1.0), the module performs time stretching on the segment: the stretching factor is set to the ratio, extending the time axis of the waveform by this factor, so that the equivalent change rate (power change amplitude divided by the stretched time) is reduced to below the threshold, while keeping the absolute amplitude of the power change unchanged. The module connects all processed waveform segments in their original time sequence to generate a corrected extreme time-series output sequence for that time scale. This adaptive processing method dynamically adjusts the correction intensity of the extreme waveform according to the real-time remaining lifespan of the electrolyzer. New equipment retains more of the original extreme fluctuations to make full use of resources, while aging equipment receives more stringent protection to extend its service life, achieving a balance between safety and economy throughout the entire life cycle of the equipment.

[0059] In some embodiments, the above-mentioned fusion of conventional time-series power output sequences at various time scales in order of time scale to obtain basic time-series power output data may further include, specifically, the wind-solar coupled power output simulation module 101 may be used to: perform layer-by-layer fusion in order of time scale from large to small: for the time scale of the current layer, its conventional time-series power output sequence is used as the reference sequence; for the time scale of the next layer, the preset components in its conventional time-series power output sequence are extracted by wavelet decomposition and amplitude calibrated according to the energy consistency principle; the calibrated preset components are superimposed on the reference sequence to generate a new reference sequence, which is used as the input for the next layer of fusion; the frequency of the preset components is greater than a preset frequency threshold; the energy consistency principle includes that the total energy of the superimposed sequence in the corresponding time window is equal to the total energy of the original high-frequency components, and the deviation from the total energy of the next layer of conventional time-series power output sequence is minimal.

[0060] The wind-solar coupled power output simulation module 101 can perform layer-by-layer fusion of conventional time-series power output sequences in descending order of time scale. The module uses the seasonal time-series power output sequence (sampling interval of months) as the initial reference sequence, which reflects the seasonal variation patterns within the year. When moving to the next layer (daily scale), the module performs discrete wavelet decomposition on the daily-scale conventional time-series power output sequence, using the Daubechies4 wavelet basis function to decompose the sequence into low-frequency approximate components and high-frequency detail components. The module extracts high-frequency detail components with frequencies greater than a preset frequency threshold (e.g., 0.5 cycles / hour), which correspond to rapid fluctuations caused by intraday weather processes. The module performs amplitude calibration on these high-frequency components, following the energy consistency principle: the total energy (sum of squared amplitudes at each moment) of the calibrated high-frequency component within each sliding time window (window length set to 7 days) is equal to the total energy of the original high-frequency component within that window, while minimizing the total energy deviation between the calibrated high-frequency component and the current reference sequence within the same time window (by solving a global scaling factor using the least squares method). The module superimposes the calibrated high-frequency components onto the current baseline sequence (seasonal-scale sequence) to generate a new sequence as input for the next layer (hourly-scale) fusion. This energy consistency calibration ensures that the fused sequence is consistent with the original sequence in terms of statistical energy characteristics, avoiding energy distortion caused by simple superposition. This allows the generated basic time-series power output data to maintain the physical characteristics of multiple time scales while possessing accurate fluctuation intensity statistics, providing a reliable input boundary for subsequent hydrogen production optimization.

[0061] In some embodiments, the above-mentioned method of superimposing the modified extreme time-series output sequences at each time scale onto the base time-series output data in reverse order of time scale to obtain wind and solar time-series simulated output data may further include, specifically, the wind-solar coupled output simulation module 101 may be used to: process the modified extreme time-series output sequences at each time scale in reverse order of time scale from smallest to largest; for the modified extreme time-series output sequence at the current time scale, obtain the state of charge range and upper limit of charging and discharging power of the energy storage unit, as well as the available volume and upper limit of hydrogen output flow of the hydrogen storage system; and, with the physical capacity of the energy storage unit and the hydrogen storage system as constraints, simulate superimposing the current modified extreme time-series output sequence onto the current base time-series output data. The process is as follows: The extreme power fluctuations at each moment are decomposed into high-frequency components that need to be instantaneously smoothed by the energy storage unit and low-frequency components that need to be buffered by the hydrogen storage system. If the remaining capacity of the energy storage unit is insufficient to smooth the high-frequency components at that moment, this high-frequency power fluctuation is decomposed and transferred to the energy storage task at the next moment. If the available volume of the hydrogen storage system is insufficient to absorb the low-frequency components, the excess power is marked as unabsorbable power. The corrected extreme time-series output sequence, verified and adjusted by energy storage and hydrogen storage capacity constraints, is superimposed onto the current basic time-series output data to update the basic time-series output data. The corrected extreme time-series output sequence for the next time scale is processed, and so on, until all time scales are processed to obtain the simulated wind and solar time-series output data.

[0062] The wind-solar coupled power output simulation module 101 can sequentially overlay the corrected extreme time-series power output sequences of each time scale onto the base time-series power output data in reverse order of time scale (e.g., from hourly to daily to seasonal). When processing the corrected extreme sequence of the current time scale, the module obtains two key parameters of the energy storage unit: the state of charge range (typically between 20% and 95% of rated capacity) and the upper limit of charge / discharge power (in megawatts), and two key parameters of the hydrogen storage system: available volume (unoccupied space in the hydrogen storage tank, in standard cubic meters) and the upper limit of hydrogen output flow (maximum amount of hydrogen released per unit time). The module simulates the overlay process as follows: For each moment in the corrected extreme sequence, the extreme fluctuation power is decomposed into two parts—the high-frequency part (fluctuation duration less than 5 minutes) needs to be instantaneously smoothed by the energy storage unit, and the low-frequency part (fluctuation duration greater than or equal to 5 minutes) needs to be buffered by the hydrogen storage system. If the current remaining capacity (in kilowatt-hours) of the energy storage unit is insufficient to mitigate the high-frequency component at a given moment, the module decomposes the excess power according to the principle of temporal proximity and transfers it to energy storage tasks at adjacent moments. For example, the excess power is evenly distributed among energy storage tasks within one hour before and after the current moment in a 1:1 ratio, without increasing the total energy storage task volume at the moment of transfer. If the available volume of the hydrogen storage system is insufficient to absorb the low-frequency component, the module marks this excess power as "unabsorbable power" and does not participate in this superposition. The module adds the corrected extreme sequence values, which have been verified and are actually superimposed, to the current base time-series output data, updating the base time-series output data. After processing the extreme sequences at the current time scale, the same operation is performed on the next, coarser time-scale extreme sequence. After processing all time scales, the final wind and solar time-series simulated output data strictly meets the physical capacity limitations of the energy storage and hydrogen storage systems, avoiding the generation of extreme output scenarios that are not feasible in engineering. This capacity constraint verification method solves the major flaw in traditional wind and solar power output simulation that ignores storage device bottlenecks, making it impossible to implement optimized configuration schemes in actual operation. It significantly improves the engineering usability of simulation data and the feasibility of planning schemes.

[0063] In some embodiments, the determination of the time-series power generation data of the wind and solar power generation unit may further include, specifically, the integrated optimization module 104 may also be used to: calculate the initial time-series power generation data based on the installed capacity of wind and solar power and the simulated output data of wind and solar power; simulate the charging and discharging strategy of the energy storage unit and the power tracking process of the hydrogen production unit based on the initial time-series power generation data; when the power change rate in the initial time-series power generation data exceeds the power tracking rate threshold of the hydrogen production unit during the simulation, determine the amount of power wasted due to peak shaving and valley filling of the excess power; and generate corrected time-series power generation data based on the initial time-series power generation data after peak shaving and valley filling and the amount of power wasted.

[0064] The integrated optimization module 104 can determine the time-series power generation data of the wind and solar power generation units. Based on the given wind and solar installed capacity, and combined with the simulated wind and solar power output data (i.e., the per-unit output curve multiplied by the installed capacity), the module calculates the total wind and solar power generation hourly to obtain the initial time-series power generation data. The module simulates the charging and discharging behavior of the energy storage unit and the power tracking process of the hydrogen production unit: the electrolyzer in the hydrogen production unit has a power tracking rate threshold, which represents the maximum allowable power change per unit time. For example, for alkaline electrolyzers, this is 5% to 10% of the rated power per minute, and for proton exchange membrane electrolyzers, it is 30% to 50% per minute. When the power change rate of adjacent hours in the initial time-series power generation data exceeds the tracking rate threshold of the hydrogen production unit, the module determines that the excess cannot be directly absorbed by the electrolyzer and needs to be peak-shaving and valley-filling by the energy storage unit. The specific operation of peak-shaving and valley-filling is as follows: the power increment exceeding the threshold is temporarily stored in the energy storage unit and released during subsequent periods when the power change rate is lower than the threshold. The module records the actual amount of electricity wasted during the process due to the efficiency loss of the energy storage unit during charging and discharging (e.g., the overall charging and discharging efficiency of lithium-ion energy storage is 85% to 92%), which is the abandoned electricity. The module outputs the time-series power data after peak shaving and valley filling as the corrected time-series power generation data. This method, by pre-simulating the effect of energy storage on mitigating extreme power fluctuations, makes the power generation data closer to the actual absorption capacity, avoiding overestimation of the system's hydrogen production potential by directly using raw wind and solar power generation data.

[0065] In some embodiments, the determination of the first system configuration scheme of the wind-solar-hydrogen storage system and the total power demand of the consumption unit based on time-series power generation data may further include, specifically, the integrated optimization module 104 may also be used to: construct a first configuration optimization model with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, and with the operating constraints of the hydrogen production unit, the charging and discharging efficiency constraints of the energy storage unit, and the three-rate coupling matching constraints as conditions; solve the first configuration optimization model based on time-series power generation data to obtain a first system configuration scheme; the first system configuration scheme includes a first number of activated electrolyzers in the hydrogen production unit, a first power allocation strategy, and a first hourly hydrogen production rate; and determine the total power demand of the consumption unit based on the first hourly hydrogen production rate of each electrolyzer.

[0066] The integrated optimization module 104 can determine the first system configuration scheme and the total power demand of the consumption unit based on time-series power generation data. The module constructs a first configuration optimization model with the dual optimization objectives of minimizing wind and solar curtailment rates and maximizing hydrogen production efficiency. Constraints include: operational constraints for the hydrogen production unit, such as the minimum stable operating power of the electrolyzer (typically 20% to 30% of rated power), minimum start-up / shutdown time interval, and upper limit of power change rate; charging and discharging efficiency constraints for the energy storage unit, such as maintaining the state of charge between 20% and 95% and ensuring charging and discharging power does not exceed rated power; and the aforementioned three-rate coupling matching constraints (dynamic matching of wind and solar power output change rate, electrolyzer load following rate, and chemical load regulation rate). The module uses time-series power generation data as known input and calls a mixed-integer linear programming solver or genetic algorithm to solve the model, obtaining the first system configuration scheme. This scheme includes the first number of electrolyzers in operation in the hydrogen production unit, the power allocation value of each electrolyzer at different times (i.e., the first power allocation strategy), and the resulting hourly hydrogen production (i.e., the first hourly hydrogen production). The module sums the hourly hydrogen production of each electrolyzer and divides it by the baseline hydrogen consumption rate of the consumption unit to convert it into the total power demand of the consumption unit under ideal steady state (expressed as hydrogen flow rate or equivalent electricity). This approach couples configuration optimization and operation strategy into the same model, avoiding suboptimal solutions where configuration is determined first and operation is determined later, thus improving the overall economic efficiency of the system design.

[0067] In some embodiments, the above-mentioned optimization objective of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, and the conditions of the operation constraints of the wind-solar-hydrogen storage system, the charge and discharge efficiency constraints of the energy storage unit, and the three-rate coupling matching constraints, are used to construct a first configuration optimization model. Specifically, the integrated optimization module 104 can also be used to: use the wind and solar time-series simulated output data as the observation state sequence, and the start-up and shutdown states and operating power levels of the electrolyzer as the hidden state sequence to construct a hidden Markov model; wherein, the state transition probability matrix is ​​non-uniformly corrected according to the start-up and shutdown loss coefficient of the electrolyzer and the shortest continuous operating time constraint, and the observation probability matrix is ​​established according to the hydrogen production efficiency curve and power tracking rate threshold of the electrolyzer in different hidden states; to maximize the probability of the hidden Markov model generating the observation sequence, and in combination with the electrolyzer configuration decision variables, the first configuration optimization model is constructed.

[0068] The integrated optimization module 104 can introduce a hidden Markov model into the first configuration optimization model to describe the dynamic behavior of the electrolyzer. The module uses the simulated wind and solar power output data as an observation state sequence, with a length of 8760 hours, where each observation is the normalized total wind and solar power value. The start-up and shutdown states and operating power levels of the electrolyzer are used as a hidden state sequence. The start-up and shutdown states include four discrete states: shutdown, startup, operation, and shutdown. The operating power level is divided into multiple intervals (e.g., 10% step size). The module constructs a state transition probability matrix, where each element represents the probability of transitioning from one hidden state to another. This probability is non-uniformly corrected based on the electrolyzer's start-up and shutdown loss coefficient (performance degradation or energy loss caused by each start-up and shutdown) and the minimum continuous operating time constraint (the electrolyzer must run for at least a certain number of hours after startup): the transition probability is reduced for high start-up and shutdown losses, and the transition probability for violating the minimum operating time is set to zero. The module constructs an observation probability matrix, where each element represents the probability of observing a specific wind and solar power value under a given hidden state. This probability is determined based on the hydrogen production efficiency curve of the electrolyzer at the corresponding operating power level (i.e., the hydrogen production per unit of electrical power as a function of load rate) and the power tracking rate threshold. If the wind and solar power changes exceed the tracking threshold, the observation probability decreases significantly. The module aims to maximize the probability of the observation sequence generated by the hidden Markov model, while introducing electrolyzer configuration decision variables (such as the number of electrolyzers and the power per unit) to construct a first configuration optimization model. This model automatically obtains the start-up and shutdown timings and power allocation strategies of the electrolyzers throughout the year through optimal inference of the hidden state sequence, while ensuring that the configuration scheme statistically matches the characteristics of wind and solar power fluctuations. This approach solves the suboptimal problem caused by the need for manually preset start-up and shutdown rules in traditional optimization, achieving adaptive generation from wind and solar power statistical characteristics to electrolyzer operating strategies.

[0069] In some embodiments, the above-mentioned calculation of the required wind and solar installed capacity based on total electricity demand and wind and solar time-series simulated power output data may further include, specifically, the integrated optimization module 104 may also be used to: construct a second configuration optimization model with the optimization objectives of minimizing wind and solar installed capacity, minimizing wind and solar curtailment rate, and maximizing hydrogen production efficiency, and with the hydrogen consumption constraints of the consumption unit and the three-rate coupling matching constraints as conditions; the hydrogen consumption constraints include the hydrogen demand range constraints and hydrogen demand adjustment rate constraints of the consumption unit; solve the second configuration optimization model to obtain the minimum wind and solar installed capacity that meets the total electricity demand and the second system configuration scheme of the wind-solar-hydrogen storage system; the second system configuration scheme includes the second number of activated electrolyzers in the hydrogen production unit, the second power allocation strategy, and the second hourly hydrogen production.

[0070] The integrated optimization module 104 can inversely determine the required wind and solar installed capacity given the total electricity demand. The module constructs a second configuration optimization model with three optimization objectives: minimizing the wind and solar installed capacity, minimizing the wind and solar curtailment rate, and maximizing hydrogen production efficiency. Constraints include hydrogen consumption constraints for the consumption unit: a hydrogen demand range constraint (the hourly hydrogen demand of the consumption unit must be between the minimum and maximum allowable values ​​of the equipment) and a hydrogen demand regulation rate constraint (the change in hydrogen demand between adjacent hours cannot exceed the upper limit of the chemical plant's regulation capacity). The module converts the total electricity demand into a time-series hydrogen demand curve for the consumption unit, using it as known input. The module uses the wind and solar installed capacity (the respective capacities of wind and solar power) and the electrolyzer configuration (number of units, power per unit) of the hydrogen production unit as decision variables, and solves for the minimum wind and solar installed capacity that satisfies the time-series hydrogen demand curve under the three-rate coupling matching constraint. The solution process employs a multi-objective evolutionary algorithm (such as NSGA-III) to simultaneously optimize the three objectives. The solution results include: the minimum wind power capacity and minimum photovoltaic capacity to meet total electricity demand (i.e., the minimum installed wind and solar capacity), as well as the second number of electrolyzers to be put into operation, the second power allocation strategy, and the second hourly hydrogen production rate under this installed capacity, collectively referred to as the second system configuration scheme. This approach achieves a precise reverse derivation from downstream electricity demand to upstream wind and solar resource allocation, avoiding the inefficiency and blindness of manual calculations, and ensuring the lowest investment cost while meeting the consumption demand.

[0071] In some embodiments, the above-mentioned optimization objectives are to minimize the installed capacity of wind and solar power, minimize the curtailment rate of wind and solar power, and maximize the hydrogen production efficiency. The second configuration optimization model is constructed based on the hydrogen consumption constraints of the consumption unit and the three-rate coupling matching constraints. Specifically, the integrated optimization module 104 can also be used to: construct a bidirectional hidden Markov model using the time-series hydrogen demand of the consumption unit as a constraint observation sequence, and the installed capacity of wind and solar power and the configuration parameters of the electrolyzer as hidden state variables. In the forward process, the hydrogen production is predicted using the time-series output of wind and solar power as input; in the backward process, the required wind and solar power output is deduced from the downstream hydrogen demand. The second configuration optimization model is constructed by fusing the probabilities of both directions, with the objectives of minimizing the error between the forward-predicted hydrogen production and the backward-deduced hydrogen demand, while simultaneously minimizing the installed capacity of wind and solar power.

[0072] The integrated optimization module 104 can construct a bidirectional hidden Markov model in the second configuration optimization model to achieve source-load coupling inversion. The module uses the time-series hydrogen demand of the absorption unit as the constraint observation sequence. The wind and solar installed capacity and electrolyzer configuration parameters are used as hidden state variables, where the wind and solar installed capacity is a continuous variable and the electrolyzer configuration is a discrete variable. The module constructs a bidirectional hidden Markov model: the forward process uses the time-series wind and solar power output (obtained by multiplying the wind and solar installed capacity by the per-unit output curve) as input, predicting hydrogen production through the electrolyzer model; the backward process uses the time-series hydrogen demand as input, inversely extrapolating the required wind and solar power output level. There is an error between the forward-predicted hydrogen production and the backward-extended hydrogen demand; the module uses the sum of squares of this error as an objective function. Simultaneously, the module uses minimizing the wind and solar installed capacity as another objective function. The module constructs a joint probability maximization model by fusing the forward probability (the probability of producing sufficient hydrogen given the wind and solar installed capacity) and the backward probability (the probability of requiring the required wind and solar installed capacity given the hydrogen demand). Solving this model can simultaneously yield the minimum installed capacity of wind and solar power that satisfies the hydrogen demand constraint, as well as the corresponding electrolyzer configuration scheme. This two-way coupling approach utilizes the statistical framework of Hidden Markov Models to uniformly represent the fluctuation characteristics of source-end resources and the characteristics of load-end demand in the same probabilistic model. This solves the problem of slow convergence due to repeated iterations caused by the lack of communication between source and load information in traditional one-way optimization, and achieves a one-time solution to the globally optimal source-load matching scheme.

[0073] In some embodiments, the steps of determining the first system configuration scheme and the first operation control strategy, and determining the second system configuration scheme and the second operation control strategy are iteratively executed until a preset termination condition is met, resulting in a third system configuration scheme and a third operation control strategy; or, the steps of determining the second system configuration scheme and the second operation control strategy, and determining the first system configuration scheme and the first operation control strategy are iteratively executed until a preset termination condition is met, resulting in a third system configuration scheme and a third operation control strategy. Specifically, this may further include: iteratively executing the above steps with the goal of minimizing the difference between the first system configuration scheme and the determined second system configuration scheme, and minimizing the difference between the first operation control strategy and the second operation control strategy, until a preset termination condition is met; and determining the third system configuration scheme and the third operation control strategy based on the first system configuration scheme, the first operation control strategy, the second system configuration scheme, and the second operation control strategy that meet the preset termination condition.

[0074] The integrated optimization module 104 performs iterative calculations between forward and reverse optimization until the results in both directions converge. Forward optimization starts with a given wind and solar installed capacity and outputs a first system configuration scheme and a first operation control strategy. The first system configuration scheme includes the number of electrolyzers in the hydrogen production unit, their rated power per unit, and the mixing power ratio; the rated power and rated capacity of the energy storage unit; and the volume and hydrogen charging / discharging rate of the hydrogen storage unit. The first operation control strategy includes hourly start / stop and power allocation of the electrolyzers, hourly charging / discharging of the energy storage, hourly charging / discharging of the hydrogen storage, and hourly load adjustment of the chemical plant. Reverse optimization starts with a given end-product output and outputs a second system configuration scheme and a second operation control strategy, with the same content dimensions as the first scheme. The module's iterative objective is to minimize the difference between the first and second system configuration schemes, and simultaneously minimize the difference between the first and second operation control strategies. The differences between configuration schemes are calculated using a weighted sum of the absolute values ​​of the relative errors of each capacity parameter, with the weights determined by the proportion of investment costs; the differences between control strategies are calculated using the root mean square error of the hourly operating variables throughout the year. After each round of forward and backward optimization, the module calculates the current bidirectional difference value. If the difference value is greater than a preset convergence threshold (e.g., capacity difference less than 5% and strategy difference less than 8%), the configuration scheme output from the previous round of forward optimization is used as a new constraint for backward optimization, while the configuration scheme output from the previous round of backward optimization is used as a new starting point for forward optimization. This process is repeated alternately before entering the next iteration. Iteration stops when both bidirectional difference values ​​are below their respective thresholds. The module takes the arithmetic mean of the results from the last round of forward and backward optimization (the capacity parameter is taken as the mean, and the control strategy is taken as the mean at different time points) as the third system configuration scheme and the third operational control strategy. This convergence condition ensures that the two logics of forward "source-determined production" and backward "production-determined source" reach consistency within the same optimization framework, avoiding scheme divergences caused by different starting points. The output configuration scheme simultaneously satisfies both installed capacity constraints and production constraints, achieving a balanced and optimal balance between the source and load sides.

[0075] In some embodiments, the calculation of the first matching degree between the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence may specifically include, that is, the electrolyzer optimized hydrogen production module 102 may also be used to: for each time point, calculate the ratio of the sum of hydrogen production, hydrogen storage release, and energy storage release to hydrogen demand; if the ratio is greater than 1, the excess is marked as surplus; if the ratio is less than 1, the difference is marked as hydrogen deficiency; calculate the cumulative proportion of the surplus and hydrogen deficiency over the total time length, and subtract the cumulative proportion from 1 to obtain the first matching degree.

[0076] The electrolyzer optimization hydrogen production module 102 can calculate the first degree of matching between the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence. The wind and solar power generation sequence is the total power output of wind and solar power generation per hour; the hydrogen production sequence is the hourly hydrogen production of the electrolyzer; the hydrogen storage sequence is the hourly net hydrogen release from the hydrogen storage tank (release is positive, storage is negative); the energy storage sequence is the hourly net discharge of electrochemical energy storage (discharge is positive, charging is negative); and the hydrogen demand sequence is the hourly hydrogen demand of downstream chemical plants. For each time point (per hour), the module calculates the sum of hydrogen production, hydrogen storage release, and energy storage release (i.e., the effective energy contributed by discharge), and then divides it by the hourly hydrogen demand to obtain a ratio. If the ratio is greater than 1, it indicates that the total amount of hydrogen production and storage release in that hour exceeds the downstream demand, and the excess is marked as surplus (unit: kilograms of hydrogen or standard cubic meters). If the ratio is less than 1, it indicates that the total amount is insufficient, and the difference is marked as hydrogen shortage (unit: same). The module iterates through 8760 hours throughout the year, counting the number of hours with excess capacity and the number of hours with hydrogen shortage, and calculates their cumulative proportion over the total time (i.e., excess hours / 8760 and hydrogen shortage hours / 8760). Subtracting the sum of these two proportions from 1 yields the first matching degree. This matching degree ranges from 0 to 1, with a higher value indicating a higher proportion that the system can precisely meet the downstream hydrogen demand in each hour. This calculation method directly reflects the temporal matching degree of energy supply and demand between the source, storage, and load, with clear physical meaning, facilitating subsequent comprehensive evaluation with the second matching degree.

[0077] In some embodiments, the calculation of the first matching degree between the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence may further include, specifically, the electrolyzer optimized hydrogen production module 102 may also be used to: obtain the upper and lower bounds of the hydrogen demand fluctuation of the consumption unit to form a hydrogen demand tolerance band; for each time point, compare the sum of hydrogen production, hydrogen storage release, and energy storage release with the hydrogen demand tolerance band; mark the time points where the sum falls within the tolerance band as matching points; and calculate the proportion of the number of matching points to the total number of time points as the first matching degree.

[0078] When calculating the first matching degree, the hydrogen production module 102 of the electrolyzer optimization can introduce upper and lower bounds for the hydrogen demand fluctuation of the consumption unit. The upper bound for the hydrogen demand fluctuation is the maximum allowable hourly hydrogen supply of the downstream chemical unit; exceeding this value will prevent the unit from operating safely. The lower bound for the hydrogen demand fluctuation is the minimum allowable hydrogen supply; below this value, the unit may shut down. The upper and lower bounds together constitute the hydrogen demand tolerance band, which takes into account the actual operational flexibility of the chemical unit. For example, a synthetic ammonia unit can operate safely within 70% to 105% of its rated load. For each time point, the module compares the sum of hydrogen production, hydrogen storage release, and energy storage discharge with the hydrogen demand tolerance band. If the sum is greater than or equal to the lower bound and less than or equal to the upper bound, the time point is marked as a matching point; if it exceeds the upper bound or falls below the lower bound, it is marked as a mismatch point. The module calculates the proportion of the number of matching points to the total number of time points (8760 hours) and uses this proportion as the first matching degree. This matching degree characterization system, after considering the actual operability range of downstream devices, represents the proportion of time during which hydrogen can be safely supplied. It is more in line with engineering practice, avoids over-configuring energy storage in pursuit of theoretically accurate matching, reduces investment costs, and ensures safe operation.

[0079] In some embodiments, the calculation of the first matching degree between the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence may further include, specifically, the electrolyzer optimized hydrogen production module 102 may also be used to: perform empirical mode decomposition on the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence respectively to obtain their respective intrinsic mode components and residual components; perform time alignment on the decomposition results; perform cross-correlation analysis on each pair of aligned intrinsic mode components to obtain the correlation coefficients between each sequence at multiple time scales; and use the weighted average of the correlation coefficients as the first matching degree, wherein the weight of low-frequency components is higher than that of high-frequency components.

[0080] The electrolyzer optimization hydrogen production module 102 can employ a multi-scale signal decomposition method when calculating the first matching degree. The module performs empirical mode decomposition (EMD) on the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence, respectively. EMD adaptively decomposes each non-stationary sequence into several intrinsic mode components (arranged from high to low frequency) and a residual component (representing the trend term). The decomposition yields the fluctuation components of each sequence at different time scales. The module then aligns the intrinsic mode components of each sequence according to the same frequency order, i.e., aligning the highest frequency components, the second highest frequency components, and so on. The module performs cross-correlation analysis on each pair of aligned intrinsic mode components (e.g., the high-frequency components of the wind and solar power generation sequence and the high-frequency components of the hydrogen demand sequence), calculates the Pearson correlation coefficient, and obtains the correlation between the sequences at that frequency scale. The module calculates a weighted average of correlation coefficients across all scales. The weighting coefficient for low-frequency components (corresponding to seasonal and monthly variations) is set to twice that of high-frequency components (corresponding to hourly and daily variations), because low-frequency mismatches are more difficult to compensate for through energy and hydrogen storage and have a greater impact on system design. This weighted average result is used as the first degree of matching. This approach can distinguish matching differences across different time scales, identifying whether the mismatch is due to long-term trends or short-term fluctuations, thereby allowing for targeted adjustments to the wind-solar power ratio or energy storage capacity.

[0081] In some embodiments, the second matching degree between the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence can specifically include, that is, the electrolyzer optimized hydrogen production module 102 can also be used to: align the three sequences in time, calculate the sum of the absolute differences at each time point; take the average of the sum of the absolute differences at all time points to obtain the average deviation; and subtract the normalized average deviation from 1 to obtain the second matching degree.

[0082] The electrolyzer optimization hydrogen production module 102 can calculate the second degree of matching between the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence. The wind and solar power output change rate sequence is the ratio of the difference in wind and solar power generation in adjacent hours to the power of the previous hour; the electrolyzer load following rate sequence is the ratio of the actual change in electrolyzer power to the change in wind and solar power; and the hydrogen demand adjustment rate sequence is the ratio of the change in hydrogen demand of downstream units to the rated hydrogen demand. All three sequences have been normalized to the same dimension (unitless percentage values). The module aligns the three sequences time-by-time and calculates the sum of the absolute differences between them at each time point: |wind and solar power change rate - electrolyzer following rate| + |electrolyzer following rate - hydrogen demand adjustment rate| + |hydrogen demand adjustment rate - wind and solar power change rate|. This sum reflects the consistency of the direction and magnitude of change of the three at the same moment. The module calculates the arithmetic mean of the sum of the absolute differences at all time points to obtain the average deviation. The average deviation is normalized by minimum-maximum (divided by the maximum possible deviation value, for example, 2) to obtain a normalized average deviation between 0 and 1. The normalized average deviation is then subtracted from 1 to obtain the second matching degree. A higher matching degree indicates that the load following behavior of the electrolyzer closely follows both wind and solar power fluctuations and matches changes in downstream hydrogen demand, resulting in better dynamic coordination among the three factors.

[0083] In some embodiments, the calculation of the second matching degree between the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence may further include, specifically, the electrolyzer optimized hydrogen production module 102 may also be used to: perform sliding window standardization on the wind and solar power output change rate sequence and the hydrogen demand adjustment rate sequence respectively to obtain the instantaneous fluctuation intensity curves of the two sequences; calculate the correlation coefficient between the electrolyzer load following rate sequence and the instantaneous fluctuation intensity curve within each sliding window; and weight the correlation coefficients of each window according to the total wind and solar energy within the window to obtain the second matching degree.

[0084] The electrolyzer optimization hydrogen production module 102 can use a sliding window method to calculate the second matching degree. The module performs sliding window standardization on the wind and solar power output change rate sequence and the hydrogen demand adjustment rate sequence respectively: setting the window length to 6 hours and the step size to 1 hour, within each window, the sequence is subtracted from the mean within the window and then divided by the standard deviation within the window to obtain the instantaneous fluctuation intensity curves of the two sequences. These curves are dimensionless and reflect the strength of the fluctuation relative to the local average level. For each sliding window, the module calculates the Pearson correlation coefficient between the electrolyzer load following rate sequence (already standardized) and the instantaneous fluctuation intensity curve. The correlation coefficient ranges from -1 to 1, with a positive value indicating that the electrolyzer's following direction is consistent with the fluctuation direction, and a negative value indicating reverse adjustment. The module calculates the total wind and solar energy within each window (i.e., the sum of wind and solar power generation in each hour within the window), and uses this total energy as a weighting coefficient to perform a weighted average of the correlation coefficients for each window. The weighted average result is used as the second matching degree. This approach divides long-term time series into multiple local intervals using a sliding window, enabling the assessment of differences in the dynamic following ability of electrolyzers under different weather conditions (such as continuous cloudy days or windy periods). The introduction of weights makes the matching quality of high-energy intervals (high output and high regulation requirements) have a greater impact on the final score, which is more in line with the engineering focus.

[0085] In some embodiments, the calculation of the second matching degree between the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence may further include, specifically, the electrolyzer optimized hydrogen production module 102 may further be used to: perform Hilbert transform on the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence respectively to obtain the instantaneous phase sequence of the three; calculate the first instantaneous phase difference between the electrolyzer load following rate sequence and the wind and solar power output change rate sequence, and the second instantaneous phase difference between the electrolyzer load following rate sequence and the hydrogen demand adjustment rate sequence; take the weighted sum of the absolute values ​​of the first instantaneous phase difference and the second instantaneous phase difference over time to obtain the average phase deviation; and subtract the normalized average phase deviation from 1 to obtain the second matching degree.

[0086] When calculating the second matching degree, the electrolyzer optimization hydrogen production module 102 can use Hilbert transform to extract the instantaneous phase of the signal. The module performs Hilbert transform on the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence, respectively. The Hilbert transform transforms the real signal into a complex analytic signal, from which the instantaneous phase angle (ranging from 0 to 360 degrees) at each time point can be extracted. The instantaneous phase reflects the periodic movement position of the signal at that moment. The module calculates the instantaneous phase difference between the electrolyzer load following rate sequence and the wind and solar power output change rate sequence (i.e., the absolute value of the phase angle difference between the two sequences at the same moment), as well as the instantaneous phase difference between the electrolyzer load following rate sequence and the hydrogen demand adjustment rate sequence. Ideally, the electrolyzer following rate should be in phase with the wind and solar power change rate (phase difference close to 0) and also in phase with the hydrogen demand adjustment rate, indicating that the electrolyzer operation is synchronized with the source load fluctuation. The module calculates the average value of the two phase differences over the whole year (arithmetic mean of the phase differences of each hour), obtaining the average phase deviation A and the average phase deviation B. The module takes the weighted sum of the two values ​​as the total average phase deviation. The weighting coefficient is set according to the sensitivity of the downstream unit to load following; for example, the weight for hydrogen demand matching is set to 0.6, and the weight for wind and solar matching is set to 0.4. After normalizing the total average phase deviation (dividing by the maximum possible phase difference of 180 degrees), the second matching degree is obtained by subtracting this value from 1. This matching degree reflects the degree of lag or lead in phase of the electrolyzer's dynamic response, and can identify tracking asynchrony problems caused by control delays or equipment inertia, providing a basis for optimizing control parameters.

[0087] In some embodiments, the first conditional distribution for fitting the hydrogen supply amount under the first load to be greater than the first hydrogen supply threshold may further include, that is, the flexible chemical consumption module 103 may further be used to: obtain all sample points for the hydrogen supply amount to be greater than the first hydrogen supply threshold under the first load, use a generalized Pareto distribution to perform maximum likelihood estimation on the excess amount of hydrogen supply for the sample points, set the shape parameter to be greater than zero, and obtain the first conditional distribution; the first conditional distribution is used to characterize the tail characteristics of the chemical product output growth rate decreasing with the increase of hydrogen supply and not approaching a fixed upper bound.

[0088] The flexible chemical processing module 103 can fit the conditional distribution when the hydrogen supply exceeds a high hydrogen supply threshold within a high-load range. The high-load range can be the operating range where the chemical plant's load rate exceeds a preset proportion (e.g., 80%), and the high hydrogen supply threshold is a preset quantile (e.g., 90%) of the hydrogen supply distribution within this range. The module acquires all sample points where the hydrogen supply exceeds the threshold within the high-load range; each sample point contains a hydrogen supply value and the corresponding chemical product output. For each sample point, the excess amount of hydrogen supply exceeding the threshold is calculated (i.e., hydrogen supply minus the threshold). The module uses a generalized Pareto distribution to perform maximum likelihood estimation on the excess amount sequence, obtaining the shape and scale parameters of the distribution. The module forces the shape parameter to be greater than zero; at this point, the distribution has an unbounded heavy-tailed characteristic, meaning that theoretically, regardless of how much the hydrogen supply increases, the output always increases, but the rate of increase gradually decreases, and it will not approach a fixed output upper limit. This distribution serves as the first conditional distribution, used to describe the output response law of the chemical plant under overload operation. By using a generalized Pareto distribution to fit extreme value samples, the shortcomings of the traditional normal distribution in underestimating the production growth potential under extreme high hydrogen supply scenarios are avoided, providing a more accurate high-load boundary prediction for system configuration.

[0089] In some embodiments, the first conditional distribution when the hydrogen supply under the first load is greater than the first hydrogen supply threshold may further include, specifically, the flexible chemical consumption module 103 may be used to: divide the load interval corresponding to the first load into multiple sub-intervals, each sub-interval corresponding to a different hydrogen supply excess level; perform kernel density estimation on the ratio of production to hydrogen supply of sample points in each sub-interval to obtain the local conditional density function of each sub-interval; and perform weighted concatenation of the local density functions of all sub-intervals in ascending order of hydrogen supply to obtain the first conditional distribution.

[0090] The flexible chemical processing module 103 can construct a first conditional distribution using a piecewise kernel density estimation method. The module divides the hydrogen supply range corresponding to the high-load interval into multiple sub-intervals, each sub-interval corresponding to a hydrogen supply excess threshold. The division method is as follows: starting from the high hydrogen supply threshold, the module divides at equal intervals according to the hydrogen supply increment, for example, every 100 standard cubic meters constitutes one sub-interval, resulting in 5 to 10 sub-intervals. For each sub-interval, the module extracts the hydrogen supply and corresponding output of all sample points within that sub-interval, calculating the ratio of output to hydrogen supply (yield per unit of hydrogen supply) for each sample point. The module uses a Gaussian kernel density estimation method to perform probability density estimation on the yield ratio within each sub-interval, obtaining the local conditional density function of that sub-interval. The bandwidth of the kernel density estimation is selected using cross-validation. The module weights and concatenates the local density functions of all sub-intervals in ascending order of hydrogen supply, with the weighting coefficient set as the proportion of the number of sample points in each sub-interval to the total number of sample points. During splicing, linear interpolation is used at the boundaries of adjacent sub-intervals to smooth the transition, forming a continuous first conditional distribution across the entire domain. This approach overcomes the problem of insufficient fitting of global parameter models to complex tail morphologies through piecewise modeling, and can adapt to the differentiated yield variation patterns exhibited by chemical plants under different overload conditions, thus improving the prediction accuracy of the model over a wide hydrogen supply range.

[0091] In some embodiments, the first conditional distribution for fitting the hydrogen supply under the first load when it is greater than the first hydrogen supply threshold may further include, specifically, the flexible chemical consumption module 103 may also be used to: obtain sample points where the hydrogen supply under the first load is greater than the first hydrogen supply threshold; perform a double logarithmic coordinate transformation on the excess hydrogen supply and the corresponding production increment; fit the transformed data points using piecewise linear regression in the double logarithmic coordinate system; identify the inflection point sequence where the slope gradually decreases from greater than 1 to less than 1; construct a piecewise power law function based on the inflection point sequence as the first conditional distribution; the first conditional distribution is used to characterize the tail characteristics of the continuous change in production growth from superlinear to sublinear.

[0092] The flexible chemical processing module 103 can construct a first conditional distribution using a piecewise power-law function in a double logarithmic coordinate system. The module acquires sample points where the hydrogen supply exceeds the high-load threshold within the high-load interval. For each sample point, it calculates two transformation quantities: the natural logarithm of the hydrogen supply excess (hydrogen supply minus the threshold), and the natural logarithm of the corresponding production increment (actual production minus the production corresponding to the threshold). The transformed data points are plotted in a double logarithmic coordinate system. The module uses a piecewise linear regression algorithm to fit these data points. The algorithm automatically searches for the optimal inflection point position, dividing the data sequence into multiple continuous line segments. Starting from a point where the hydrogen supply excess is relatively small, it identifies an inflection point sequence where the slope gradually decreases from greater than 1 to less than 1. A slope greater than 1 indicates that the production growth rate is faster than the hydrogen supply growth rate (superlinear), a slope less than 1 indicates that the production growth rate is slower than the hydrogen supply growth rate (sublinear), and a slope equal to 1 indicates linear growth. The module constructs piecewise power-law functions based on the inflection point sequence: within each segment interval, the increase in production and the excess amount of hydrogen supply satisfy a power-law relationship, with the power exponent equal to the slope of the segment. The piecewise power-law functions of all segments are combined in ascending order of excess hydrogen supply to form a globally continuous first conditional distribution. This distribution can characterize the tail characteristics of the continuous change in production growth from superlinear (due to scale effects) to sublinear (due to mass transfer limitations). Compared to a single generalized Pareto distribution, it provides a more detailed characterization of growth patterns, helping to accurately capture the marginal revenue change patterns under high loads during optimal allocation.

[0093] In some embodiments, the above-mentioned determination of the second conditional distribution of hydrogen supply under the second load based on historical hydrogen consumption rate may further include, specifically, the flexible chemical consumption module 103 may also be used to: project the hydrogen supply-production data points under the second load onto a feature space with hydrogen consumption rate as the variable, cluster the hydrogen consumption rate distribution using a Gaussian mixture model to obtain multiple hydrogen consumption rate sub-modes; fit a linear sub-model to each sub-mode and describe the transition probability between sub-modes using a hidden Markov model; and fuse each linear sub-model according to the steady-state probability of the hidden Markov model to obtain the second conditional distribution.

[0094] The flexible chemical processing module 103 can determine the linear mapping relationship between hydrogen supply and product output within the second load range. The second load range can be the steady-state operating range of the chemical plant with a load rate between 30% and 80%. The module acquires all historical data points within this range, with each data point containing the hydrogen supply and the corresponding product output. The module projects each data point onto a feature space with hydrogen consumption rate as the variable, where hydrogen consumption rate equals the hydrogen supply divided by the product output, in standard cubic meters per ton. The module uses a Gaussian mixture model to cluster the hydrogen consumption rate distribution, assuming that the data is composed of a weighted combination of multiple Gaussian distributions. The module estimates the number of Gaussian components (2 to 4) and the mean, variance, and mixing weight of each component using the expectation-maximization algorithm. Each Gaussian component corresponds to a hydrogen consumption rate sub-mode, such as a low hydrogen consumption mode (high-efficiency zone), a medium hydrogen consumption mode (normal zone), and a high hydrogen consumption mode (low-efficiency zone). For each sub-mode, the module fits a linear sub-model, in the form that output equals hydrogen supply divided by the average hydrogen consumption rate of that sub-mode. This linear sub-model reflects the baseline characteristic that output increases linearly with hydrogen supply under that sub-mode. The module further constructs a Hidden Markov Model (HMM) to describe the transition probabilities between sub-modes. The hidden state of the HMM is the current hydrogen consumption rate sub-mode, and the observed value is the actual hydrogen consumption rate. The module statistically analyzes the frequency of transitions from one sub-mode to another from historical data, using this as elements of the state transition probability matrix. The module calculates the steady-state probability distribution of the HMM, i.e., the long-term proportion of each sub-mode occurring over a long period of operation. The module weights and fuses the linear sub-models according to their steady-state probabilities to obtain a second conditional distribution. This fusion method, through a probabilistic framework, handles the random fluctuations in hydrogen consumption rate of the chemical plant under different operating conditions, enabling the linear baseline model to adaptively reflect the long-term average efficiency of the plant and avoiding the systematic bias caused by a single fixed hydrogen consumption rate.

[0095] In some embodiments, the above-mentioned determination of the second conditional distribution of hydrogen supply under the second load based on historical hydrogen consumption rate may further include, that is, the above-mentioned flexible chemical consumption module 103 may further be used to: acquire historical hydrogen consumption rate data under different ambient temperatures under the second load, establish a correction function between ambient temperature and hydrogen consumption rate; divide the benchmark hydrogen consumption rate by the ambient temperature correction coefficient to obtain the dynamic benchmark hydrogen consumption rate; and construct the second conditional distribution that is adjusted in real time with the ambient temperature using the dynamic benchmark hydrogen consumption rate as the slope.

[0096] The flexible chemical processing module 103 can construct a second conditional distribution that adjusts in real time according to ambient temperature. The module acquires historical hydrogen consumption rate data under different ambient temperatures within the second load range. The ambient temperature data comes from meteorological station records at the project site, ranging from -30℃ to +40℃. The module analyzes the relationship between hydrogen consumption rate and ambient temperature, finding that decreased catalytic activity at low temperatures leads to an increased hydrogen consumption rate, while equilibrium constraints at high temperatures also cause an increase. An optimal temperature range (e.g., 15℃ to 25℃) exists where the hydrogen consumption rate is lowest. The module uses a quadratic polynomial or spline interpolation to fit a correction function between ambient temperature and hydrogen consumption rate. This function uses ambient temperature as the independent variable and outputs a temperature correction coefficient, defined as the ratio of the hydrogen consumption rate at that temperature to the hydrogen consumption rate at a baseline temperature (e.g., 20℃). The module obtains a fixed baseline hydrogen consumption rate, which is taken from the design value of the unit under standard operating conditions (20℃, standard atmospheric pressure) or the median of the hydrogen consumption rate from historical best operating data. For a given real-time ambient temperature, the module calculates the corresponding temperature correction coefficient. The baseline hydrogen consumption rate is then divided by this correction coefficient (or multiplied by its reciprocal) to obtain the dynamic baseline hydrogen consumption rate. Using the reciprocal of the dynamic baseline hydrogen consumption rate as the slope, the module constructs a linear function: output equals hydrogen supply divided by the dynamic baseline hydrogen consumption rate. This function serves as a second conditional distribution that adjusts in real-time according to ambient temperature. This approach quantifies the impact of ambient temperature on chemical reaction efficiency and embeds it into the absorption model. This allows the mapping relationship between hydrogen supply and output to automatically adapt to temperature changes when planning in different seasons and climate regions, improving the model's geographical applicability and year-round prediction accuracy.

[0097] In some embodiments, the aforementioned fitting of the third conditional distribution when the hydrogen supply under the third load is less than the second hydrogen supply threshold may further include, specifically, the aforementioned flexible chemical consumption module 103 may further be used to: obtain all sample points where the hydrogen supply under the third load is less than the second hydrogen supply threshold, fit the deviation of the hydrogen supply below the threshold using a generalized Pareto distribution, set the shape parameter to be less than zero, and obtain the third conditional distribution; the third conditional distribution is used to characterize the tail characteristics of the production rate decreasing faster and having a minimum operable production limit when the hydrogen supply decreases.

[0098] The flexible chemical processing module 103 can fit the conditional distribution when the hydrogen supply is lower than the low hydrogen supply threshold within a low-load range. The low-load range can be the operating range where the chemical plant's load rate is less than a preset proportion (e.g., 30%), and the low hydrogen supply threshold is a preset quantile (e.g., 10%) of the hydrogen supply distribution within this range. The module acquires all sample points where the hydrogen supply is lower than the threshold within the low-load range. Each sample point contains a hydrogen supply value and the corresponding chemical product output. For each sample point, the deviation of the hydrogen supply below the threshold (i.e., the threshold minus the hydrogen supply) is calculated. The module uses a generalized Pareto distribution to perform maximum likelihood estimation on the deviation sequence, obtaining the shape and scale parameters of the distribution. The module forces the shape parameter to be less than zero. At this point, the distribution has a heavy-tailed characteristic with an upper bound, meaning that after the hydrogen supply decreases to a certain level, the rate of output decline accelerates sharply (manifested as the conditional probability density function tends to infinity at the boundary), and there exists a minimum operable output limit (this limit is determined by the minimum hydrogen supply required for the chemical plant to maintain basic reactions; below this value, the plant must shut down). This distribution, as a third conditional distribution, is used to describe the production decline of chemical plants under low-load and insufficient hydrogen supply conditions. By fitting the low-side extrema with a generalized Pareto distribution with a negative shape parameter, the physical constraints of accelerated production decline and hard boundaries can be accurately captured, providing a safe operating boundary for system configuration under low hydrogen supply risk.

[0099] In some embodiments, the above-mentioned fitting of the third conditional distribution when the hydrogen supply under the third load is less than the second hydrogen supply threshold may further include, that is, the above-mentioned flexible chemical consumption module 103 may further be used to: calculate the deviation of the hydrogen supply and the rate of decrease in production for each sample point under the third load; fit the relationship between the deviation and the rate of decrease using an exponential decay model; and use the exponential decay model as the third conditional distribution.

[0100] The flexible chemical processing module 103 can construct a third conditional distribution using an exponential decay model. The module acquires all sample points within the low-load zone where the hydrogen supply is less than the low hydrogen supply threshold. For each sample point, two characteristic quantities are calculated: hydrogen supply deviation, calculated as (low hydrogen supply threshold minus actual hydrogen supply) divided by the low hydrogen supply threshold (divided dimensionlessly, ranging from 0 to 1); and production decline rate, calculated as (benchmark production minus actual production) divided by the benchmark production, where the benchmark production is the expected production corresponding to the low hydrogen supply threshold (obtained through extrapolation from a linear model in the medium-load zone). The module uses the hydrogen supply deviation as the independent variable and the production decline rate as the dependent variable, fitting the model using an exponential decay model. The model form is: decline rate equals 1 minus e raised to the power of negative k times the deviation, where k is the decay coefficient, determined by the least squares method. This model describes the exponential growth characteristic of production decline, which initially slows down and then accelerates as the hydrogen supply deviation increases, consistent with the physical mechanism of decreased catalyst activity and increased heat loss ratio under low load conditions. The module uses the exponential decay model as the third conditional distribution and outputs the expected rate of production decline for a given deviation in hydrogen supply.

[0101] In some embodiments, the above-mentioned fitting of the third conditional distribution when the hydrogen supply under the third load is less than the second hydrogen supply threshold may further include, that is, the above-mentioned flexible chemical consumption module 103 may further be used to: obtain sample points where the hydrogen supply under the third load is less than the second hydrogen supply threshold, divide the process of hydrogen supply being lower than the threshold into multiple consecutive low hydrogen supply events; extract three features for each event: minimum hydrogen supply, low hydrogen supply duration, and minimum production valley value; fit the joint distribution of the above features using a three-dimensional Cauchy distribution, with the marginal distribution set to a thick-tailed shape, to obtain the third conditional distribution.

[0102] The flexible chemical processing module 103 can use a three-dimensional Cauchy distribution to perform joint probability modeling of continuous low hydrogen supply events. The module acquires all continuous time segments within the low-load interval where the hydrogen supply is less than the low hydrogen supply threshold, defining each segment as a continuous low hydrogen supply event. For each event, the module extracts three features: the lowest hydrogen supply value in the event (in standard cubic meters per hour), the duration of the event (in hours, the time from the first time the hydrogen supply falls below the threshold to the last time it falls below the threshold), and the lowest production trough during the event (in tons per hour, taking the minimum production value within the event). The module collects the three-dimensional feature vectors of all events and uses a three-dimensional Cauchy distribution for joint probability density fitting. The three-dimensional Cauchy distribution has a thick-tailed marginal distribution shape, and its probability density function decays according to a power law at the tail (the decay rate is slower than that of a Gaussian distribution), which can better adapt to the rare scenarios of extreme low values, long durations, and deep production troughs in low hydrogen supply events. The module sets the scale parameters and correlation matrices of the three marginal distributions, determines the parameter values ​​through maximum likelihood estimation, and obtains the third conditional distribution. This distribution can be used to assess the probability and severity of a system falling into a severely low-production state under a given hydrogen supply fluctuation pattern, providing a risk assessment basis for energy storage and hydrogen storage capacity planning under extreme operating conditions. This approach, through event-level joint modeling, differs from traditional point-by-point independent modeling, preserving the temporal correlation information of the low hydrogen supply process and more accurately reflecting the cumulative effect of chemical plants under continuous insufficient hydrogen supply.

[0103] In some embodiments, the wind-solar coupled hydrogen production and consumption integrated optimization system can adopt a B / S architecture, which includes a physical layer, a data hub layer, a model engine layer, a back-end service layer, and a front-end interaction layer from the front end to the back end and from the bottom up.

[0104] The physical layer is the platform's hardware infrastructure, employing a dual-operating system deployment mode of "Linux + Windows". The Linux system server cluster hosts the task engine, computing services, and distributed storage; while the Windows system servers support the model development environment and high-precision mathematical programming solutions.

[0105] The data hub layer is configured as a distributed storage cluster (including a business storage database cluster and a massive data storage cluster) to build and store a multi-dimensional collaborative resource database and define standardized data interfaces to realize data flow and verification between layers.

[0106] Database content: The database must integrate and store at least the following heterogeneous data from multiple sources: Wind and solar resource data: including global atmospheric reanalysis data such as ERA5 and MERRA2, mesoscale meteorological data, meteorological station data, and wind tower measured data.

[0107] Equipment parameter data: including mainstream wind turbine models and power curves, photovoltaic module models and conversion efficiencies, PEM / ALK electrolyzer technical parameters, and energy storage system technical parameters.

[0108] Downstream process data includes mathematical model parameters for the synthesis processes of green ammonia, green alcohol, and green aviation fuel, hydrogen consumption rates, load adjustment ranges, and production start-up and shutdown response times; as well as typical biomass components and carbon source data for the Northeast and Northwest regions.

[0109] Geographic and meteorological data: including the latitude and longitude of the project location, climate characteristics, and extreme weather event data.

[0110] Data Interfaces: To achieve efficient collaboration between modules, the data hub layer defines and opens four types of standardized data interfaces: The first type of interface (wind and solar-hydrogen production interface) connects the wind and solar resource model with the electrolyzer optimization model. It is used to transmit hourly output data of wind and solar power after cleaning and interpolation (such as the 8960-hour power curve) and resource volatility.

[0111] The second type of interface (hydrogen production-hydrogen consumption interface): connects the electrolyzer optimization model and the flexible chemical conversion model, and is used to transmit parameters such as hydrogen production, electrolyzer operating efficiency, and start-up / shutdown status.

[0112] The third type of interface (wind-solar-hydrogen direct connection interface): connects the wind-solar resource model and the flexible chemical conversion model, and is used to directly transmit core indicators such as total installed capacity of wind and solar power, theoretical power generation, and curtailment rate in specific optimization scenarios.

[0113] The fourth type of interface (unified output interface) is used to format (JSON / XML) the results of each model's calculations (such as equipment capacity, cost indicators, and energy balance sheets), synchronize them to the front-end interaction layer for visualization, and store them in the data central layer for archiving.

[0114] The model engine layer is connected to the data hub layer and is configured under the Windows operating system environment. It integrates multiple core planning modules. These models are called by the backend service layer through standardized interfaces, forming the computing core of the platform. It includes a wind-solar coupled power output simulation module 101, an electrolyzer optimized hydrogen production module 102, a flexible chemical processing module 103, and an integrated optimization module 104.

[0115] The model engine layer also includes an off-grid energy storage adaptation model: This model is specifically designed for off-grid wind-solar coupled hydrogen production projects, focusing on three mainstream wind turbine types: doubly-fed asynchronous generators, direct-drive permanent magnet synchronous generators, and semi-direct-drive permanent magnet synchronous generators. Through electromagnetic transient simulation and steady-state operation analysis, it quantifies the key parameter requirements for grid-connected energy storage under different wind and solar installed capacities, such as capacity, response speed, and charging / discharging power. It also analyzes the impact of hydrogen production load fluctuations on grid stability, ultimately outputting energy storage configuration schemes and parameter standards that balance stability and economy.

[0116] The backend service layer is configured in a Linux operating system environment and connects to the data hub layer and model engine layer respectively. It uses the Flask framework to build a RESTful API and is the core for handling business logic and task scheduling.

[0117] Task scheduling module: Based on the Celery distributed task scheduler and Supervisor daemon, it implements queued and asynchronous processing of coupled computing tasks. After a user submits a task, the system automatically assigns it to an idle computing node, provides real-time feedback on the task status, and triggers a retry mechanism in case of failure.

[0118] Business logic module: responsible for parsing front-end requests, performing project management, access control (based on RBAC model), parameter validation, and coordinating the sequential calling of various models in the model engine layer (for example, calling the wind and light model first, and its output as the input of the electrolytic cell model, and so on), and finally processing the calculation results and returning them in a formatted manner.

[0119] The front-end interaction layer connects to the back-end service layer and is developed based on modern web frameworks such as Vue.js. It provides an intuitive and easy-to-use user interface for project and task management, parameter configuration, task initiation, and visualization of planning results. Specifically, it includes: System access module: Enables user login, permission verification, password modification, and secure logout.

[0120] The system homepage module serves as the core carrier of "data dashboard + function navigation," displaying the platform's core indicators, statistical analysis charts, and function entry points.

[0121] Project Management Module: Enables full lifecycle management (add, delete, modify, and query) of project information and supports the association and binding of "project-task".

[0122] Task Management Module: As the core scheduling carrier for the platform's full-process computing tasks, it integrates data acquisition, parameter configuration, and multi-model collaborative computing functions to achieve full-process integration and closed-loop of "data-parameter-computation-result".

[0123] Model Calculation Module: Serving as the interaction entry point between users and the underlying professional models, this module encapsulates five core models (wind and solar power output, electrolyzer optimization, industrial hydrogen production, ammonia and methanol production, and grid-type energy storage), providing independent parameter configuration and result viewing interfaces.

[0124] System management module: For system administrators, it provides role management, permission management, department management, user management, log management and system monitoring functions to ensure the safe and stable operation of the platform.

[0125] As can be seen from the technical solutions provided in the embodiments of this specification above, the embodiments of this specification can generate wind and solar time-series simulated power output data based on multi-source wind and solar resource data; based on the wind and solar time-series simulated power output data, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, and under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints, an electrolyzer hydrogen production scheme is determined; a consumption model is constructed to characterize the mapping relationship between hydrogen supply and chemical product output under multiple loads; based on the electrolyzer hydrogen production scheme and consumption model, the system configuration scheme and operation control strategy are optimized according to the wind and solar installed capacity or the target chemical product output. By using a three-rate coupling matching constraint, the power output of wind and solar power, the load of electrolyzers, and the chemical industry consumption are dynamically coordinated to avoid power curtailment or frequent equipment start-ups and shutdowns caused by fluctuations and mismatches, thereby improving the stability of the power generation and hydrogen production sides. Furthermore, based on the mapping relationship between hydrogen supply and product output under various load conditions, the adaptability of the chemical industry side to fluctuating hydrogen sources is enhanced. The integrated two-way optimization flexibly configures the system according to the installed capacity of wind and solar power or the target output, realizing the integrated coordinated optimization of power generation, hydrogen production, and consumption.

[0126] The following are three specific embodiments of this specification: Implementation 1: Implementation of the overall platform architecture.

[0127] This embodiment provides a wind-solar coupled hydrogen production integrated planning and design platform based on full-chain collaborative optimization. Its physical layer employs a high-performance server cluster. One or more Linux servers (Ubuntu 20.04) serve as core business servers, deploying Nginx 1.20 for load balancing and running Flask applications, Celery task queues, and Redis caching. One or more Windows Server 2019 servers serve as model computation servers, running Python 3.8 and Matlab R2021a, and integrating mathematical programming solvers such as Gurobi 9.5 to support the core algorithm calculations of the model engine layer. The data hub layer uses a MySQL 8.0 cluster (master-slave architecture) to store business data, combined with a MongoDB cluster to store massive amounts of time-series wind measurement data and calculation results.

[0128] Implementation 2: Collaborative computation implementation at the model engine layer.

[0129] When a user initiates a "reverse optimization" calculation task in the task management module, with the goal of "producing 100,000 tons of green ammonia annually, requiring a unit cost of less than 3,000 yuan / ton", the collaborative workflow of the model engine layer is as follows: The flexible chemical processing module is activated first. Based on the annual production target of 100,000 tons of green ammonia, and combined with its internally integrated Haber-Bosch process mathematical model, it deduces the required stable and reliable annual hydrogen supply demand (approximately 18,000 tons) and the hourly hydrogen supply fluctuation tolerance range.

[0130] This hydrogen supply demand is transmitted to the electrolyzer optimization hydrogen production module through a second interface. The model combines typical annual meteorological data from the wind and solar resource database, with the goal of minimizing hydrogen supply costs, and preliminarily solves the required wind and solar installed capacity range, energy storage ratio, and hybrid configuration scheme of PEM / ALK electrolyzers while meeting hourly hydrogen supply demand.

[0131] Wind-solar coupled output simulation module: Based on the initial wind and solar capacity range given by the electrolyzer model, it performs more refined micro-site selection simulation in the specified area, generates accurate hourly wind and solar output curves, evaluates the actual wind and solar curtailment rate, and feeds this data back to the electrolyzer model through the first interface.

[0132] The integrated optimization module receives data from all modules and constructs a full-system nonlinear mathematical model, including power balance, hydrogen balance, equipment capacity limitations, investment costs, operation and maintenance costs, and product revenue. It solves the mixed-integer nonlinear programming problem by calling the Gurobi solver, iteratively outputting the optimal solution that satisfies the requirement of "annual production of 100,000 tons of green ammonia with a cost ≤ 3000 yuan". For example, a hybrid solution might include 300MW wind power, 200MW photovoltaic power, a supporting 40MW / 80MWh electrochemical energy storage system, and four 1000Nm³ / h ALK electrolyzers and two 500Nm³ / h PEM electrolyzers, along with a detailed annual scheduling strategy.

[0133] Implementation 3: Implementation of the off-grid-to-grid energy storage adaptation model.

[0134] In an off-grid wind-solar-hydrogen production project, the user selected three doubly-fed asynchronous wind turbines of a certain model (3MW each) and a fixed-capacity photovoltaic array. The user inputs the wind and solar scale through the off-grid model interface in the front-end interaction layer. The back-end service layer calls the off-grid energy storage adaptation model. This model first simulates the start-up characteristics and fault response characteristics of the wind turbines without grid support based on a built-in electromagnetic transient simulation algorithm. Then, combining the fluctuation probability model of wind and solar power output and the step change scenario of downstream hydrogen production load, it calculates the required energy storage support power to meet the system's frequency stability (e.g., frequency deviation not exceeding ±0.5Hz), voltage stability, and black start requirements. Through multi-constraint weighted summation, the model ultimately calculates that, in this scenario, to ensure stable off-grid operation, the minimum power of the energy storage system required is 8MW. Based on the 2-hour standby time standard, the minimum capacity should be 16MWh, and this is output as the recommended configuration. The results are ultimately displayed to the user in a clear chart format on the front-end page.

[0135] Based on the aforementioned integrated optimization system for wind-solar coupled hydrogen production and consumption, this specification also proposes embodiments of the integrated optimization method for wind-solar coupled hydrogen production and consumption. For example... Figure 2 As shown, the integrated optimization method for wind-solar coupled hydrogen production and consumption may specifically include: S201: Generate time-series simulated power output data for wind and solar resources based on multi-source wind and solar resource data; S202: Based on the simulated output data of wind and solar power, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, the electrolyzer hydrogen production scheme is determined under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints. S203: Construct a consumption model to characterize the mapping relationship between hydrogen supply and chemical product output under various load conditions; S204: Based on the electrolyzer hydrogen production scheme and consumption model, optimize the system configuration scheme and operation control strategy according to the wind and solar power installed capacity or the target chemical product output.

[0136] In some embodiments, the above-mentioned optimization of system configuration and operation control strategies based on electrolyzer hydrogen production schemes and consumption models, according to the scale of wind and solar power capacity or target chemical product output, includes: If the installed capacity of wind and solar power is given, the time-series power generation data of the wind and solar power generation units are determined based on the given installed capacity and the time-series simulated power output data; based on the time-series power generation data, the first system configuration scheme, the first operation control strategy, and the total power demand of the consumption unit are determined. Based on the total electricity demand and the wind and solar power output data from the time-series simulation, the required wind and solar installed capacity, the second system configuration scheme, and the second operation control strategy are derived. The inverted wind and solar installed capacity is used as the new given wind and solar installed capacity. Based on the new given wind and solar installed capacity, the steps of determining the first system configuration scheme and the first operation control strategy, and determining the second system configuration scheme and the second operation control strategy are iteratively executed until the preset termination condition is met, and the third system configuration scheme and the third operation control strategy are obtained.

[0137] In some embodiments, the above-mentioned optimization of system configuration and operation control strategies based on the electrolyzer hydrogen production scheme and consumption model, according to the wind and solar power installed capacity or the target chemical product output, further includes: If the installed capacity of wind and solar power is not given, determine the total power demand of the consumption unit based on the output of the target chemical products; based on the total power demand and the time-series simulated output data of wind and solar power, inversely derive the required installed capacity of wind and solar power, the configuration scheme of the second system, and the second operation control strategy. Based on the inverted wind and solar installed capacity and the wind and solar time-series simulated output data, the time-series power generation data of the wind and solar power generation units are determined; based on the time-series power generation data, the first system configuration scheme, the first operation control strategy, and the total power consumption demand of the consumption unit are determined. The determined total electricity demand is taken as the new total electricity demand; based on the new total electricity demand, the steps of determining the second system configuration scheme and the second operation control strategy, as well as determining the first system configuration scheme and the first operation control strategy, are iteratively executed until the preset termination condition is met, and the third system configuration scheme and the third operation control strategy are obtained.

[0138] In some embodiments, determining the time-series power generation data of the wind and solar power generation unit includes: Calculate the initial time-series power generation data based on the installed capacity of wind and solar power and the time-series simulated power output data of wind and solar power. Based on initial time-series power generation data, the charging and discharging strategies of the energy storage unit and the power following process of the hydrogen production unit are simulated. When the power change rate in the initial time-series power generation data exceeds the power tracking rate threshold of the hydrogen production unit during the simulation, the amount of power wasted due to peak shaving and valley filling is determined. Based on the initial time-series power generation data and the amount of power wasted after peak shaving and valley filling, the corrected time-series power generation data is generated.

[0139] In some embodiments, determining the first system configuration scheme, the first operation control strategy, and the total power demand of the absorption unit based on time-series power generation data includes: With the optimization objectives of minimizing wind and solar curtailment rates and maximizing hydrogen production efficiency, and with the operational constraints of the hydrogen production unit, the charging and discharging efficiency constraints of the energy storage unit, and the three-rate coupling matching constraints as conditions, a first configuration optimization model is constructed. Based on time-series power generation data, a first configuration optimization model is solved to obtain a first system configuration scheme; the first system configuration scheme includes the first number of electrolyzers in operation of the hydrogen production unit, a first power allocation strategy, and a first hourly hydrogen production rate. The total power demand of the consumption unit is determined based on the first hourly hydrogen production of each electrolyzer.

[0140] In some embodiments, the above-mentioned method of inverting the required installed capacity of wind and solar power, the second system configuration scheme, and the second operation control strategy based on total electricity demand and wind and solar time-series simulated power output data includes: With the optimization objectives of minimizing the installed capacity of wind and solar power, minimizing the curtailment rate of wind and solar power, and maximizing the hydrogen production efficiency, and with the hydrogen consumption constraints of the consumption unit and the three-rate coupling matching constraints as conditions, a second configuration optimization model is constructed; the hydrogen consumption constraints include the hydrogen demand range constraints and the hydrogen demand adjustment rate constraints of the consumption unit. Solve the second configuration optimization model to obtain the minimum wind and solar installed capacity and the second system configuration scheme that meet the total power demand; the second system configuration scheme includes the second number of electrolyzers in operation of the hydrogen production unit, the second power allocation strategy, and the second hourly hydrogen production.

[0141] In some embodiments, generating wind and solar time-series simulated power output data based on multi-source wind and solar resource data includes: The multi-source wind and solar resource data are sampled at multiple preset time scales; Waveform clustering was performed on the sampling results at each time scale; The clustering results at different time scales are fused to generate the wind and solar time-series simulated power output data.

[0142] In some embodiments, the above-mentioned sampling of the multi-source wind and solar resource data at multiple preset time scales includes: The first response time constant of the PEM electrolyzer and the second response time constant of the ALK electrolyzer in the hydrogen production unit were obtained. A first sampling time scale is determined based on a first response time constant, and a second sampling time scale is determined based on a second response time constant; the first sampling time scale is smaller than the second sampling time scale. Multi-source wind and solar resource data are sampled at the first sampling time scale and the second sampling time scale respectively to obtain a first sampling sequence for characterizing high-frequency fluctuation components and a second sampling sequence for characterizing low-frequency fluctuation components.

[0143] In some embodiments, the above-mentioned waveform clustering of sampling results at each time scale includes: Based on the hydrogen demand change rate threshold of the consumption unit, the power fluctuation tolerance threshold of the wind and solar power generation unit is determined. Based on the rate of power change between adjacent sampling times in each time scale downsampled sequence, the time scale downsampled sequence is divided into normal fluctuation sequence and extreme fluctuation sequence according to the power fluctuation tolerance threshold. Waveform clustering was performed on multiple waveform sequences included in the normal fluctuation sequence and the extreme fluctuation sequence to obtain multiple classes of normal waveform sequences and multiple classes of extreme waveform sequences.

[0144] In some embodiments, the above-mentioned fusion of clustering results at various time scales to generate the wind and solar time-series simulated power output data includes: By splicing together multiple types of conventional waveform sequences at each time scale in chronological order, a conventional time-series power output sequence at that time scale is obtained. Based on the power fluctuation tolerance threshold, amplitude scaling or time stretching is performed on multiple extreme waveform sequences at each time scale to obtain the corrected extreme time series output sequence at that time scale. The conventional time-series power output sequences at various time scales are fused layer by layer according to the time scale order to obtain the basic time-series power output data. The modified extreme time series output sequences at each time scale are superimposed onto the basic time series output data in reverse order according to the time scale to obtain the wind and solar time series simulated output data.

[0145] In some embodiments, the above method further includes: Acquire meteorological grid data; the meteorological grid data includes first meteorological grid data and second meteorological grid data; Based on the measured meteorological time-series data corresponding to the first meteorological grid data, the first meteorological grid data is corrected to obtain the third meteorological grid data; the measured meteorological time-series data includes measured wind speed time-series data and measured solar radiation time-series data. For each second meteorological grid, select the fourth meteorological grids that are closest to the second meteorological grid from among the multiple third meteorological grids; The weighting coefficients for each fourth meteorological grid are determined based on the distance between each fourth meteorological grid and the second meteorological grid; the weighting coefficients are negatively correlated with the distance. Based on the meteorological values ​​and weighting coefficients of multiple fourth meteorological grids, the meteorological values ​​of the second meteorological grid are corrected to obtain the data of the fifth meteorological grid; Multi-source wind and solar resource data were determined based on data from the third and fifth meteorological grids.

[0146] In some embodiments, the above-mentioned method of correcting the first meteorological grid data based on the measured meteorological time-series data corresponding to the first meteorological grid data to obtain the third meteorological grid data includes: The meteorological deviation at each time point is obtained by comparing the data from the first meteorological grid with the measured meteorological time series data. A time window is constructed based on the meteorological fluctuation index of the region where the first meteorological grid data is located; the length of the time window is positively correlated with the meteorological fluctuation index. The meteorological deviation index is obtained by fitting the meteorological deviation at each time point within the time window. The meteorological deviation index is superimposed on the first meteorological grid data to obtain the third meteorological grid data.

[0147] In some embodiments, based on the simulated power output data of wind and solar power generation, and with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, the above-mentioned method determines the electrolyzer hydrogen production scheme under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate satisfy the three-rate coupling matching constraint, including: Based on the wind and solar power output time-series simulation data, the wind and solar power output change rate sequence, the electrolyzer load following rate sequence of the hydrogen production unit, and the hydrogen demand adjustment rate sequence of the consumption unit are determined. Based on the wind and solar power output time-series simulation data, the wind and solar power output change rate sequence, the wind and solar power generation sequence of the wind and solar power generation unit, the hydrogen production sequence of the hydrogen production unit and the load following rate sequence of the electrolyzer, the hydrogen storage sequence of the hydrogen storage system, the energy storage sequence of the energy storage unit, the hydrogen demand sequence of the consumption unit and the hydrogen demand regulation rate sequence are determined. Calculate the first degree of matching among the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence; Calculate the second degree of matching among the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence; The third matching degree is determined based on the first and second matching degrees; With the optimization objectives of minimizing wind and solar curtailment rates and maximizing hydrogen production efficiency, an electrolyzer hydrogen production scheme is determined under the condition that the third matching degree satisfies the three-rate coupling matching constraint.

[0148] In some embodiments, the above-described absorption model for characterizing the mapping relationship between hydrogen supply and chemical product output under various load conditions includes: Fit the first conditional distribution when the hydrogen supply is greater than the first hydrogen supply threshold under the first load; the first conditional distribution is used to characterize the non-linear growth of chemical product output with a slowing trend and no upper bound tail characteristic. Based on historical hydrogen consumption rates, a second conditional distribution of hydrogen supply under the second load is determined; the second conditional distribution is used to characterize the baseline characteristics of linear growth in chemical product output. The third conditional distribution is fitted when the hydrogen supply is less than the second hydrogen supply threshold under the third load; the third conditional distribution is used to characterize the tail characteristics of chemical product output, which has an upper bound and a non-linear, steep decline. Based on the first conditional distribution, the second conditional distribution, and the third conditional distribution, an absorption model is constructed. Among them, the first load is greater than the second load, and the second load is greater than the third load.

[0149] As can be seen from the technical solutions provided in the embodiments of this specification above, the embodiments of this specification can generate wind and solar time-series simulated power output data based on multi-source wind and solar resource data; based on the wind and solar time-series simulated power output data, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, and under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints, an electrolyzer hydrogen production scheme is determined; a consumption model is constructed to characterize the mapping relationship between hydrogen supply and chemical product output under multiple loads; based on the electrolyzer hydrogen production scheme and consumption model, the system configuration scheme and operation control strategy are optimized according to the wind and solar installed capacity or the target chemical product output. By using a three-rate coupling matching constraint, the power output of wind and solar power, the load of electrolyzers, and the chemical industry consumption are dynamically coordinated to avoid power curtailment or frequent equipment start-ups and shutdowns caused by fluctuations and mismatches, thereby improving the stability of the power generation and hydrogen production sides. Furthermore, based on the mapping relationship between hydrogen supply and product output under various load conditions, the adaptability of the chemical industry side to fluctuating hydrogen sources is enhanced. The integrated two-way optimization flexibly configures the system according to the installed capacity of wind and solar power or the target output, realizing the integrated coordinated optimization of power generation, hydrogen production, and consumption.

[0150] This specification also provides a computer device for an integrated optimization method of wind and solar coupled hydrogen production and consumption, including a processor and a memory for storing processor-executable instructions. Specifically, the processor can perform the following tasks according to the instructions: generating simulated wind and solar power output data based on multi-source wind and solar resource data; determining an electrolyzer hydrogen production scheme based on the simulated wind and solar power output data, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate satisfy the three-rate coupling matching constraint; constructing a consumption model to characterize the mapping relationship between hydrogen supply and chemical product output under various load conditions; and optimizing the system configuration scheme and operation control strategy based on the electrolyzer hydrogen production scheme and consumption model, according to the wind and solar installed capacity or the target chemical product output.

[0151] To execute the above instructions more accurately, please refer to... Figure 3 As shown in the embodiments of this specification, another specific computer device 300 is also provided, wherein the computer device 300 includes a network communication port 301, a processor 302 and a memory 303, and the above structures are connected by internal cables so that the various structures can perform specific data interaction.

[0152] The processor 302 can be specifically used for: generating wind and solar time-series simulated power output data based on multi-source wind and solar resource data; determining the electrolyzer hydrogen production scheme based on the wind and solar time-series simulated power output data, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints; constructing a consumption model to characterize the mapping relationship between hydrogen supply and chemical product output under multiple loads; and optimizing the system configuration scheme and operation control strategy based on the electrolyzer hydrogen production scheme and consumption model, according to the wind and solar installed capacity or target chemical product output.

[0153] The memory 303 can be used to store the corresponding instruction program.

[0154] In this embodiment, the network communication port 301 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.

[0155] In this embodiment, the processor 302 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.

[0156] In this embodiment, the memory 303 includes volatile memory and non-volatile memory. The memory 303 can include multiple layers. In digital systems, anything that can store binary data can be a memory; in integrated circuits, 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.

[0157] Furthermore, embodiments of this specification also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described... Figure 1 The instructions for the integrated optimization method of wind and solar coupled hydrogen production and consumption are shown.

[0158] Furthermore, embodiments of this specification provide a computer program product comprising a computer program that, when executed by a processor, implements the above-described... Figure 1 The method for integrated wind-solar coupled hydrogen production and consumption optimization is shown.

[0159] It should be understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this specification.

[0160] It should also be understood that, in the embodiments of this specification, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this specification generally indicates that the preceding and following related objects have an "or" relationship.

[0161] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0162] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0163] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0164] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational tasks to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The task is a function specified in one or more boxes.

[0165] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An integrated optimized system for wind-solar coupled hydrogen production and consumption, characterized in that, include: The wind-solar coupled output simulation module is used to generate wind-solar time-series simulated output data based on multi-source wind and solar resource data; The electrolyzer optimization hydrogen production module is used to determine the electrolyzer hydrogen production scheme based on the wind and solar power output data simulated over time, with the optimization objectives of minimizing the wind and solar power curtailment rate and maximizing the hydrogen production efficiency. Under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints. The flexible chemical processing module is used to construct a processing model that characterizes the mapping relationship between hydrogen supply and chemical product output under various load conditions. The integrated optimization module is used to optimize system configuration and operation control strategies based on the electrolyzer hydrogen production scheme and consumption model, according to the scale of wind and solar power installations or the target chemical product output.

2. The system according to claim 1, characterized in that, The integrated optimization module is used for: If the installed capacity of wind and solar power is given, the time-series power generation data of the wind and solar power generation units are determined based on the given installed capacity and the time-series simulated power output data; based on the time-series power generation data, the first system configuration scheme, the first operation control strategy, and the total power demand of the consumption unit are determined. Based on the total electricity demand and the wind and solar power output data from the time-series simulation, the required wind and solar installed capacity, the second system configuration scheme, and the second operation control strategy are derived. The inverted wind and solar installed capacity is used as the new given wind and solar installed capacity. Based on the new given wind and solar installed capacity, the steps of determining the first system configuration scheme and the first operation control strategy, and determining the second system configuration scheme and the second operation control strategy are iteratively executed until the preset termination condition is met, and the third system configuration scheme and the third operation control strategy are obtained.

3. The system according to claim 1, characterized in that, The integrated optimization module is also used for: If the installed capacity of wind and solar power is not given, determine the total power demand of the consumption unit based on the output of the target chemical products; based on the total power demand and the time-series simulated output data of wind and solar power, inversely derive the required installed capacity of wind and solar power, the configuration scheme of the second system, and the second operation control strategy. Based on the inverted wind and solar installed capacity and the wind and solar time-series simulated output data, the time-series power generation data of the wind and solar power generation units are determined; based on the time-series power generation data, the first system configuration scheme, the first operation control strategy, and the total power consumption demand of the consumption unit are determined. The determined total electricity demand is taken as the new total electricity demand; based on the new total electricity demand, the steps of determining the second system configuration scheme and the second operation control strategy, as well as determining the first system configuration scheme and the first operation control strategy, are iteratively executed until the preset termination condition is met, and the third system configuration scheme and the third operation control strategy are obtained.

4. The system according to claim 2 or 3, characterized in that, The integrated optimization module is also used for: Calculate the initial time-series power generation data based on the installed capacity of wind and solar power and the time-series simulated power output data of wind and solar power. Based on initial time-series power generation data, the charging and discharging strategies of the energy storage unit and the power following process of the hydrogen production unit are simulated. When the power change rate in the initial time-series power generation data exceeds the power tracking rate threshold of the hydrogen production unit during the simulation, the amount of power wasted due to peak shaving and valley filling is determined. Based on the initial time-series power generation data and the amount of power wasted after peak shaving and valley filling, the corrected time-series power generation data is generated.

5. The system according to claim 2 or 3, characterized in that, The integrated optimization module is also used for: With the optimization objectives of minimizing wind and solar curtailment rates and maximizing hydrogen production efficiency, and with the operational constraints of the hydrogen production unit, the charging and discharging efficiency constraints of the energy storage unit, and the three-rate coupling matching constraints as conditions, a first configuration optimization model is constructed. Based on time-series power generation data, a first configuration optimization model is solved to obtain a first system configuration scheme; the first system configuration scheme includes the first number of electrolyzers in operation of the hydrogen production unit, a first power allocation strategy, and a first hourly hydrogen production rate. The total power demand of the consumption unit is determined based on the first hourly hydrogen production of each electrolyzer.

6. The system according to claim 2 or 3, characterized in that, The integrated optimization module is also used for: With the optimization objectives of minimizing the installed capacity of wind and solar power, minimizing the curtailment rate of wind and solar power, and maximizing the hydrogen production efficiency, and with the hydrogen consumption constraints of the consumption unit and the three-rate coupling matching constraints as conditions, a second configuration optimization model is constructed; the hydrogen consumption constraints include the hydrogen demand range constraints and the hydrogen demand adjustment rate constraints of the consumption unit. Solve the second configuration optimization model to obtain the minimum wind and solar installed capacity and the second system configuration scheme that meet the total power demand; the second system configuration scheme includes the second number of electrolyzers in operation of the hydrogen production unit, the second power allocation strategy, and the second hourly hydrogen production.

7. The system according to claim 1, characterized in that, The wind-solar coupling output simulation module is used for: The multi-source wind and solar resource data are sampled at multiple preset time scales; Waveform clustering was performed on the sampling results at each time scale; The clustering results at different time scales are fused to generate the wind and solar time-series simulated power output data.

8. The system according to claim 7, characterized in that, The wind-solar coupling output simulation module is also used for: The first response time constant of the PEM electrolyzer and the second response time constant of the ALK electrolyzer in the hydrogen production unit were obtained. A first sampling time scale is determined based on a first response time constant, and a second sampling time scale is determined based on a second response time constant; the first sampling time scale is smaller than the second sampling time scale. Multi-source wind and solar resource data are sampled at the first sampling time scale and the second sampling time scale respectively to obtain a first sampling sequence for characterizing high-frequency fluctuation components and a second sampling sequence for characterizing low-frequency fluctuation components.

9. The system according to claim 7, characterized in that, The wind-solar coupling output simulation module is also used for: Based on the hydrogen demand change rate threshold of the consumption unit, the power fluctuation tolerance threshold of the wind and solar power generation unit is determined. Based on the rate of power change between adjacent sampling times in each time scale downsampled sequence, the time scale downsampled sequence is divided into normal fluctuation sequence and extreme fluctuation sequence according to the power fluctuation tolerance threshold. Waveform clustering was performed on multiple waveform sequences included in the normal fluctuation sequence and the extreme fluctuation sequence to obtain multiple classes of normal waveform sequences and multiple classes of extreme waveform sequences.

10. The system according to claim 9, characterized in that, The wind-solar coupling output simulation module is also used for: By splicing together multiple types of conventional waveform sequences at each time scale in chronological order, a conventional time-series power output sequence at that time scale is obtained. Based on the power fluctuation tolerance threshold, amplitude scaling or time stretching is performed on multiple extreme waveform sequences at each time scale to obtain the corrected extreme time series output sequence at that time scale. The conventional time-series power output sequences at various time scales are fused layer by layer according to the time scale order to obtain the basic time-series power output data. The modified extreme time series output sequences at each time scale are superimposed onto the basic time series output data in reverse order according to the time scale to obtain the wind and solar time series simulated output data.

11. The system according to claim 9, characterized in that, The wind-solar coupling output simulation module is also used for: Acquire meteorological grid data; the meteorological grid data includes first meteorological grid data and second meteorological grid data; Based on the measured meteorological time-series data corresponding to the first meteorological grid data, the first meteorological grid data is corrected to obtain the third meteorological grid data; the measured meteorological time-series data includes measured wind speed time-series data and measured solar radiation time-series data. For each second meteorological grid, select the fourth meteorological grids that are closest to the second meteorological grid from among the multiple third meteorological grids; The weighting coefficients for each fourth meteorological grid are determined based on the distance between each fourth meteorological grid and the second meteorological grid. The weighting coefficients are negatively correlated with distance; Based on the meteorological values ​​and weighting coefficients of multiple fourth meteorological grids, the meteorological values ​​of the second meteorological grid are corrected to obtain the data of the fifth meteorological grid; Multi-source wind and solar resource data were determined based on data from the third and fifth meteorological grids.

12. The system according to claim 11, characterized in that, The wind-solar coupling output simulation module is also used for: The meteorological deviation at each time point is obtained by comparing the data from the first meteorological grid with the measured meteorological time series data. A time window is constructed based on the meteorological fluctuation index of the region where the first meteorological grid data is located; the length of the time window is positively correlated with the meteorological fluctuation index. The meteorological deviation index is obtained by fitting the meteorological deviation at each time point within the time window. The meteorological deviation index is superimposed on the first meteorological grid data to obtain the third meteorological grid data.

13. The system according to claim 1, characterized in that, The electrolyzer-optimized hydrogen production module is used for: Based on the wind and solar power output time-series simulation data, the wind and solar power output change rate sequence, the electrolyzer load following rate sequence of the hydrogen production unit, and the hydrogen demand adjustment rate sequence of the consumption unit are determined. Based on the wind and solar power output time-series simulation data, the wind and solar power output change rate sequence, the wind and solar power generation sequence of the wind and solar power generation unit, the hydrogen production sequence of the hydrogen production unit and the load following rate sequence of the electrolyzer, the hydrogen storage sequence of the hydrogen storage system, the energy storage sequence of the energy storage unit, the hydrogen demand sequence of the consumption unit and the hydrogen demand regulation rate sequence are determined. Calculate the first degree of matching among the wind and solar power generation sequence, hydrogen production sequence, hydrogen storage sequence, energy storage sequence, and hydrogen demand sequence; Calculate the second degree of matching among the wind and solar power output change rate sequence, the electrolyzer load following rate sequence, and the hydrogen demand adjustment rate sequence; The third matching degree is determined based on the first and second matching degrees; With the optimization objectives of minimizing wind and solar curtailment rates and maximizing hydrogen production efficiency, an electrolyzer hydrogen production scheme is determined under the condition that the third matching degree satisfies the three-rate coupling matching constraint.

14. The system according to claim 1, characterized in that, The flexible chemical waste disposal module is used for: Fit the first conditional distribution when the hydrogen supply is greater than the first hydrogen supply threshold under the first load; the first conditional distribution is used to characterize the non-linear growth of chemical product output with a slowing trend and no upper bound tail characteristic. Based on historical hydrogen consumption rates, determine the second conditional distribution of hydrogen supply under the second load. The second conditional distribution is used to characterize the baseline properties of linear growth in chemical product output; The third conditional distribution is fitted when the hydrogen supply is less than the second hydrogen supply threshold under the third load; the third conditional distribution is used to characterize the tail characteristics of chemical product output, which has an upper bound and a non-linear, steep decline. Based on the first conditional distribution, the second conditional distribution, and the third conditional distribution, an absorption model is constructed. Among them, the first load is greater than the second load, and the second load is greater than the third load.

15. An optimized method for integrated wind-solar coupled hydrogen production and consumption, characterized in that, include: Based on multi-source wind and solar resource data, wind and solar time-series simulated power output data is generated; Based on the simulated output data of wind and solar power, with the optimization objectives of minimizing wind and solar curtailment rate and maximizing hydrogen production efficiency, the electrolyzer hydrogen production scheme is determined under the condition that the wind and solar power output change rate, electrolyzer load following rate, and chemical load adjustment rate meet the three-rate coupling matching constraints. Construct a consumption model to characterize the mapping relationship between hydrogen supply and chemical product output under various load conditions; Based on the electrolyzer hydrogen production scheme and consumption model, the system configuration scheme and operation control strategy are optimized according to the wind and solar power installed capacity or the target chemical product output.

Images