Hierarchical elastic constraint optimization system and method for off-grid liquid hydrogen production system

Abstract

This invention relates to the field of renewable energy hydrogen production scheduling technology, specifically to a hierarchical elastic constraint optimization system and method for off-grid liquid hydrogen production systems. The scheme first collects wind and solar power output and electrical load data to predict the system's operating mode. Then, it constructs a mixed-integer linear programming model, setting hard constraints on liquid hydrogen production and soft constraints on electrical load, and integrating multiple constraints such as liquefaction reflux, equipment mutual exclusion, and ramp-up rate. After solving, it outputs a full-time scheduling scheme. This invention solves the problems of unstable liquid hydrogen production, insufficient modeling accuracy, and non-optimal scheduling schemes in existing technologies, achieving continuous and stable liquid hydrogen production under off-grid conditions. The system operates safely and economically, and can effectively adapt to fluctuating renewable energy scenarios.

Description

Technical Field

[0001] This invention relates to the field of renewable energy hydrogen production scheduling technology, and more specifically, to a hierarchical elastic constraint optimization system and method for off-grid liquid hydrogen production systems. Background Technology

[0002] Off-grid renewable energy hydrogen production and liquid hydrogen storage and transportation technologies are gradually moving towards large-scale application. Liquid hydrogen is the core mode of long-distance hydrogen energy transportation, but existing off-grid hydrogen production scheduling schemes generally treat electricity load as a rigid constraint, making liquid hydrogen production susceptible to fluctuations in wind and solar power, and difficult to achieve stable production levels. At the same time, traditional hydrogen balance models do not consider the material closed loop of liquefaction reflux, resulting in errors in hydrogen storage state estimation; most schemes use genetic algorithms and pure linear programming, which cannot effectively handle discrete logical constraints such as equipment start-up and shutdown, and hydrogen charging and discharging mutual exclusion, resulting in poor practical executability of the scheduling schemes. In addition, existing systems lack reasonable load elasticity mechanisms and equipment safety constraints, making them prone to production interruptions when green electricity is scarce, and also posing safety hazards such as frequent equipment start-up and shutdown and sudden changes in operating conditions, failing to meet the actual needs of continuous, efficient, and stable liquid hydrogen production in off-grid scenarios. Summary of the Invention

[0003] In view of this, in order to address the shortcomings of the prior art, the present invention proposes a hierarchical elastic constraint optimization system and method for off-grid liquid hydrogen production systems, aiming to solve at least one of the problems mentioned in the background art.

[0004] In a first aspect, the present invention provides a hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system, comprising: a data acquisition module for acquiring the available wind power output sequence, available photovoltaic power output sequence, and local power load demand sequence for each time period within the scheduling cycle; The optimized control module is used to compare the total available output of wind power and photovoltaic power with the corresponding electrical load on a time-by-time basis, and to predict whether the current operating mode is excess hydrogen production mode or insufficient discharge mode. The configuration module is used to construct a mixed integer linear programming model with the optimization objective of minimizing the total operation and maintenance cost of the scheduling cycle. In the model, the liquid hydrogen storage tank's inventory at the end of the scheduling cycle is equal to the preset daily liquid hydrogen production target, which is set as the first-level mandatory hard constraint. The unmet electrical load variable is introduced into the electrical balance equation, and the local electrical load is set as the second-level optional soft constraint. A penalty cost term is added to the unmet electrical load in the model objective function. The output module is used to solve the constructed mixed-integer linear programming model and output the operating status, power allocation and media flow scheduling scheme of all devices in each time period within the scheduling cycle.

[0005] In some embodiments, the mixed-integer linear programming model includes a dynamic equation for the high-pressure hydrogen storage tank inventory: ; The equation contains a liquefaction reflux term. The liquefaction reflux term characterizes the mass flow rate of gaseous hydrogen that is fed into the liquid hydrogen liquefaction unit but has not been liquefied and is refluxed back to the high-pressure hydrogen storage tank.

[0006] In some embodiments, the objective function of the mixed-integer linear programming model includes an additional liquefaction reflux and restorage operation and maintenance cost item. .

[0007] In some embodiments, the mixed-integer linear programming model is configured with hydrogen charging / discharging mutual exclusion constraints and minimum start-up constraints for the liquefaction unit, and introduces binary variables of the hydrogen charging state. Binary variable of hydrogen release state and impose constraints ; Introducing binary variables for liquefaction device switches and impose constraints The liquefaction unit is limited to operating only within the range of rated minimum to maximum output.

[0008] In some embodiments, the mixed-integer linear programming model is further configured with external hydrogen purchase conditional coupling constraints, introducing auxiliary binary variables. ; When the high-pressure hydrogen storage tank level is higher than the safety lower limit, forced ; When the high-pressure hydrogen storage tank is below the safety limit, the purchase of hydrogen from external sources is permitted.

[0009] In some embodiments, the mixed integer linear programming model is configured with asymmetric elastic mode switching constraints, wherein when the total available power of wind and solar power is greater than the electrical load, it is an excess hydrogen production mode, and when the total available power of wind and solar power is less than the electrical load, it is an under-discharge mode. When the system is in excess hydrogen production mode, constraints are applied. =0 forces the fuel cell to shut down; When the system is in under-discharge mode, apply constraints. =0 Forced shutdown of the electrolyzer, and no change to the media handling rate of the liquid hydrogen liquefaction unit Set a mandatory shutdown constraint.

[0010] In some embodiments, the mixed-integer linear programming model is further configured with a device ramp-up rate constraint, specifically: Electrolytic cell power constraints: ; Hydrogen consumption constraints for fuel cells: ; Processing rate constraints of liquid hydrogen liquefaction units: ; in , , These are the upper limits of ramp rates for the electrolyzer, fuel cell, and liquid hydrogen liquefaction unit, respectively.

[0011] In some embodiments, the objective function expression is: The objective function includes ten cost items in sequence: wind power operation and maintenance cost, photovoltaic operation and maintenance cost, electrolyzer operation and maintenance cost, fuel cell operation and maintenance cost, hydrogen storage tank charging and discharging operation and maintenance cost, liquefaction reflux and re-storage cost, liquefaction unit operation and maintenance cost, wind and solar curtailment penalty cost, purchased hydrogen procurement cost, and electricity load unmet penalty cost.

[0012] In some embodiments, the electrical load in the objective function does not meet the penalty coefficient. The value is greater than the unit price of purchased hydrogen. ; In the model's execution logic, the mandatory hard constraint corresponding to the daily liquid hydrogen production has the highest priority. When there is a shortage of green electricity, the system prioritizes making up for the power shortage by purchasing hydrogen from outside, and then reduces the local power load. Throughout the process, the system prioritizes ensuring that the daily liquid hydrogen production target is achieved.

[0013] Secondly, this invention provides a hierarchical elastic constraint optimization method for an off-grid liquid hydrogen production system, comprising the following steps: Collect the available wind power output sequence, available photovoltaic power output sequence, and local electricity load demand sequence for each time period within the scheduling cycle; The total available output of wind power and photovoltaic power is compared with the corresponding electricity load for each time period to predict whether the current operating mode is excess hydrogen production mode or insufficient discharge mode. A mixed-integer linear programming model is constructed with the optimization objective of minimizing the total operation and maintenance cost of the scheduling cycle. In the model, the liquid hydrogen storage tank's inventory at the end of the scheduling cycle is equal to the preset daily liquid hydrogen production target, which is set as the first-level mandatory hard constraint. An unmet electrical load variable is introduced into the electrical balance equation, and the local electrical load is set as the second-level optional soft constraint. A penalty cost term is added to the unmet electrical load in the model objective function. Solve the constructed mixed-integer linear programming model to output the operating status, power allocation, and media flow scheduling scheme of all devices in each time period within the scheduling cycle.

[0014] Compared with existing technologies, the advantages of this invention are as follows: This invention sets hierarchical elastic constraints, prioritizing liquid hydrogen production as the highest-priority hard constraint. This automatically ensures "hydrogen supply first" during energy fluctuations, fundamentally guaranteeing stable liquid hydrogen production. By introducing a liquefaction reflux term to construct a material closed-loop model, errors in hydrogen storage state estimation are eliminated, improving scheduling accuracy. Adaptive control of purchased hydrogen and hydrogen charging / discharging logic is achieved using binary variables and large M constraints, balancing operational economy and hydrogen supply security. This scheme uses mixed-integer linear programming modeling, which can obtain a globally optimal solution, superior to traditional heuristic algorithms. The asymmetric mode switching design allows the liquefaction unit to operate continuously during power shortages, ensuring production continuity; the accompanying equipment ramp-up rate constraints effectively prevent damage to equipment from drastic changes in operating conditions. The overall system is adaptable to off-grid conditions with significant fluctuations in wind and solar power output, exhibiting strong robustness, high renewable energy utilization, and outstanding engineering application value.

[0015] The above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.

[0016] Other features and aspects of this disclosure will become clearer from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a functional block diagram of a hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system provided in an embodiment of the present invention. Figure 2 A flowchart illustrating the hierarchical elastic constraint optimization method for an off-grid liquid hydrogen production system provided in this embodiment of the invention. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0020] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0021] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0022] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0023] First embodiment, see reference Figure 1-2 As shown, a hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system according to an embodiment of this application is characterized by comprising: The data acquisition module is used to collect the available wind power output sequence, available photovoltaic power output sequence, and local electricity load demand sequence for each time period within the scheduling cycle. The optimized control module is used to compare the total available output of wind power and photovoltaic power with the corresponding electrical load on a time-by-time basis, and to predict whether the current operating mode is excess hydrogen production mode or insufficient discharge mode. The configuration module is used to construct a mixed integer linear programming model with the optimization objective of minimizing the total operation and maintenance cost of the scheduling cycle. In the model, the liquid hydrogen storage tank's inventory at the end of the scheduling cycle is equal to the preset daily liquid hydrogen production target, which is set as the first-level mandatory hard constraint. The unmet electrical load variable is introduced into the electrical balance equation, and the local electrical load is set as the second-level optional soft constraint. A penalty cost term is added to the unmet electrical load in the model objective function. The output module is used to solve the constructed mixed-integer linear programming model and output the operating status, power allocation and media flow scheduling scheme of all devices in each time period within the scheduling cycle.

[0024] It should be understood that the system is a dispatching optimization control system supporting off-grid liquid hydrogen production, which relies on wind power and photovoltaic power, two types of renewable energy, as the sole power source, and is used for scenarios such as remote areas and offshore platforms that are not supported by large power grids. The acquisition module is responsible for real-time acquiring and sorting the wind power available output sequence, photovoltaic available output sequence and local electrical load demand sequence within a 24-hour dispatching cycle and at a single-hour time step, and the data sampling frequency is consistent with the dispatching step; the optimal control module receives the data source from the acquisition module, and numerically compares the total available wind-photovoltaic output and real-time electrical load period by period to divide the system operation conditions; the configuration module builds an optimization model based on a mixed integer linear programming algorithm, establishes a two-level constraint architecture, sets the daily liquid hydrogen output as an unbreakable mandatory hard constraint, meanwhile introduces an unmet electrical load variable to convert the local electrical load into an elastically adjustable soft constraint, and adds a corresponding penalty cost item to the objective function; after completing the model solution, the output module uniformly outputs dispatching instructions such as equipment operation status, power distribution, and hydrogen medium flow in each period.

[0025] Each module of the system is linked sequentially and operates collaboratively, forming a closed-loop control from the data input end to the instruction output end, and can complete full-automatic dispatching without manual intervention. For example, in an off-grid hydrogen production station in the northwest Gobi, the four modules operate continuously to complete data acquisition, mode judgment, model calculation and instruction issuance continuously throughout the day.

[0026] The present technical solution realizes the integrated intelligent dispatching of the off-grid liquid hydrogen production system, the modular design makes the system function division clear, and later maintenance and upgrading more convenient; the standardized data acquisition and working condition prediction logic can accurately capture the output fluctuation characteristics of renewable energy; the model configuration method with hierarchical constraints guarantees the priority of core production objectives from the architecture; the fully automatic instruction output mode greatly reduces the on-site manual operation intensity. Meanwhile, the whole system has strong compatibility, can be adapted to wind power and photovoltaic off-grid hydrogen production stations with different installed capacities, and effectively improves the automation level and overall operation efficiency of the off-grid liquid hydrogen production system.

[0027] In some specific embodiments, the mixed integer linear programming model is provided with a high-pressure hydrogen tank inventory dynamic equation: ; The equation contains a liquefaction reflux term , which represents the mass flow of gaseous hydrogen that is sent to the liquid hydrogen liquefaction device and not liquefied and refluxed to the high-pressure hydrogen tank.

[0028] is the dispatching period index, is the stock of the high-pressure hydrogen tank in period t, is the hydrogen discharge flow of the hydrogen tank in period t, is the liquefaction efficiency of the liquefaction device, is the hydrogen processing flow rate of the liquefaction device at time t.

[0029] It should be understood that the high-pressure hydrogen storage module matched with the system is composed of a compressor, a pressure reducing valve and a high-pressure storage tank with a capacity range of 500kg to 5000kg. The liquid hydrogen liquefaction device adopts Claude liquefaction cycle, and the liquefaction efficiency is fixed at 0.393, which means that only 39.3% of the hydrogen fed into the liquefaction device is liquefied into liquid hydrogen, and the remaining 60.7% of gaseous hydrogen cannot be liquefied and will flow back to the high-pressure hydrogen storage tank through a special return pipeline. The liquefaction return term is an accurate quantification of this physical process. During system operation, as the core buffer unit for gaseous hydrogen, the high-pressure hydrogen storage tank integrates three hydrogen sources including hydrogen produced by electrolysis, hydrogen released for energy supply and liquefied return hydrogen, and strictly follows the inventory equation to achieve mass conservation.

[0030] For example, when the liquefaction device operates at rated load, a large amount of non-liquefied gaseous hydrogen continuously flows back to the high-pressure storage tank, and the inventory dynamic equation can calculate the storage capacity change of the storage tank in real time. This technical feature completely overcomes the defect of unconserved hydrogen material caused by traditional models that ignore liquefaction return, accurately restores the real physical process of the liquefaction link, and greatly reduces the estimation deviation of the storage capacity of the high-pressure hydrogen storage tank; based on accurate hydrogen balance modeling, the scientificity of subsequent scheduling decisions is significantly improved, which avoids problems such as equipment start-stop errors and hydrogen supply interruption caused by misjudgment of hydrogen storage status; meanwhile, this modeling method fits the actual working conditions of industrial sites, makes the simulation model highly consistent with the operating status of physical equipment, improves the matching degree of implementation of scheduling schemes, and also provides a reliable data basis for subsequent whole-system cost accounting and working condition prediction.

[0031] The electric power balance constraint is the key carrier for realizing elastic scheduling in the present method. In each time period, the sum of actual wind power output, actual photovoltaic output and fuel cell power generation is equal to the sum of electric load demand (minus the unmet electric load), power consumption of the electrolyzer and power consumption of the liquefaction device: In each time period, the sum of the actual wind power output, actual photovoltaic output and fuel cell power generation is equal to "electric load demand (minus unmet demand) + power consumption of the electrolyzer + power consumption of the liquefaction device"; wherein is the unmet electric load. It is this variable that converts the electric load from a traditional rigid constraint to an elastically adjustable soft constraint, and is the core carrier of elastic scheduling of the present invention; The hydrogen storage capacity of the high-pressure hydrogen storage tank maintains dynamic balance during the charge and discharge cycle. In the cross-time scheduling model, the storage capacity of the storage tank is a state variable, which is jointly determined by the hydrogen filling amount, hydrogen release amount and liquefaction return amount, and its mass conservation constraint relationship is as follows: wherein Defined as liquefaction reflux, its physical meaning is the amount of gaseous hydrogen fed into the liquefaction unit but not liquefied (proportion: The mass flow rate of the high-pressure hydrogen storage tank is returned. The introduction of this term eliminates the mass non-conservation problem in traditional models that treat hydrogen fed into the liquefaction unit as "direct loss," achieving a closed-loop material mass balance in the liquefaction process within the current optimization framework. The inventory dynamics of the liquid hydrogen storage tank are relatively simple, determined only by the successfully liquefied portion, where the state equation can be expressed as: For all Established.

[0032] In some specific embodiments, the objective function of the mixed-integer linear programming model includes an additional term for the operation and maintenance cost of liquefaction reflux and restorage. .

[0033] This is the operation and maintenance cost coefficient per unit mass of hydrogen in the high-pressure hydrogen storage unit.

[0034] It should be understood that, among them The unit price for the re-storage and maintenance of hydrogen per unit mass in the high-pressure hydrogen storage unit is a fixed parameter derived from on-site operation and maintenance costs such as equipment depreciation, energy consumption, and labor. Considering the liquefaction reflux physical process, the gaseous hydrogen flowing back from the liquefaction unit to the high-pressure hydrogen storage tank needs to undergo further compression and tank pressure stabilization to complete storage. This process incurs additional operation and maintenance costs, which are specifically quantified in this cost item. During model operation, this cost item, along with other cost items such as wind power, photovoltaics, and electrolyzers, participates in the calculation of the total operation and maintenance cost, becoming an important component of the optimization objective.

[0035] For example, when the liquefaction unit operates at high load for extended periods and the liquefaction reflux rate is large, this cost will increase accordingly. The model will automatically optimize the operating load of the liquefaction unit based on this cost, balancing production capacity and maintenance expenses. Adding this cost item enables complete cost accounting for the entire liquid hydrogen production chain, addressing the shortcoming of missing reflux hydrogen storage costs in traditional solutions, and ensuring the cost model fully reflects actual on-site expenses. During cost optimization, the model can comprehensively consider the relationship between liquefaction capacity, reflux scale, and storage costs, avoiding uncontrolled maintenance costs due to blindly increasing the liquefaction load. Simultaneously, the complete cost system facilitates economic analysis, electricity price and hydrogen price calculations at the site, helping off-grid liquid hydrogen sites achieve refined operation and management, further improving the overall economic efficiency and operational control capabilities of the system.

[0036] In some specific embodiments, the mixed-integer linear programming model is configured with hydrogen charging / discharging mutual exclusion constraints and minimum start-up constraints for the liquefaction unit, and introduces binary variables of the hydrogen charging state. Binary variable of hydrogen release state and impose constraints ; Introducing binary variables for liquefaction device switches and impose constraints The liquefaction unit is limited to operating only within the range of rated minimum to maximum output.

[0037] / / =Binary variable corresponding to the device status (0 = off / 1 = running); =Minimum start-up flow rate of liquefaction unit =Maximum rated flow rate of the liquefaction unit.

[0038] It should be understood that high-pressure hydrogen storage tanks are equipped with two sets of pipelines for hydrogen filling and discharging. The physical structure of the tank dictates that hydrogen filling and discharging operations cannot be performed simultaneously. Therefore, a binary variable for hydrogen filling is introduced. Hydrogen release binary variable ,pass Constraints enforce mutual exclusion of operations; liquefaction units are set with a rated minimum output. With maximum rated output Combined with switch binary variables Range constraints are constructed to limit the liquefaction unit to complete shutdown or stable operation within the range of minimum to maximum output. In engineering practice, the minimum output of the liquefaction unit is generally set at 30% of the rated capacity to avoid unstable operation under low load. During system operation, hydrogen charging and discharging mutual exclusion constraints monitor the status of the high-pressure storage tank pipeline in real time, while minimum start-up constraints limit the operating range of the liquefaction unit, preventing the equipment from starting, stopping, or operating at extremely low flow rates.

[0039] For example, when the high-pressure storage tank is in the hydrogen charging state, the system automatically locks the hydrogen release pipeline; when the liquefaction unit is started, the load will always be maintained above the minimum output. This set of constraints first avoids the physical operation conflict of simultaneous charging and releasing of hydrogen in the high-pressure hydrogen storage tank, preventing safety accidents such as pipeline pressure disorder and equipment damage; secondly, it effectively avoids frequent start-up and shutdown of the liquefaction unit and low-load operation, reduces mechanical fatigue and performance degradation of cryogenic liquefaction equipment, and significantly extends the service life of the liquefaction unit and high-pressure hydrogen storage equipment; at the same time, the discrete variable constraints make the model more accurate in describing the equipment operation logic, improve the executability of the scheduling scheme on the physical equipment, and ensure the long-term, stable and safe operation of the entire hydrogen system.

[0040] In some specific embodiments, the mixed-integer linear programming model is further configured with external hydrogen purchase conditional coupling constraints, introducing auxiliary binary variables. ; When the high-pressure hydrogen storage tank level is higher than the safety lower limit, forced ; When the high-pressure hydrogen storage tank is below the safety limit, the purchase of hydrogen from external sources is permitted.

[0041] = External hydrogen purchase prohibited state binary variable, M = large M constant used in mixed integer programming. =Lower limit of safe stock level for hydrogen storage tanks =Time period for purchased hydrogen flow.

[0042] It should be understood that when the real-time inventory level in the storage tank is higher than the safety lower limit, the constraint is mandatory. This would lock down the external hydrogen purchase channel and prohibit external hydrogen purchases; only when the storage tank inventory falls below the safety threshold and self-produced green hydrogen cannot meet the system's needs will such purchases be permitted. A value of 0 automatically unlocks the external hydrogen purchase channel, enabling emergency use of an external hydrogen source. During system operation, two constraint formulas work together to determine the storage tank's inventory status and dynamically switch between enabling and disabling external hydrogen purchases, requiring no manual intervention throughout the process.

[0043] For example, when wind and solar power output is normal and storage tank inventory is sufficient, the external hydrogen purchase channel remains closed; when encountering continuous wind and solar power outages and inventory shortages, the system automatically activates external hydrogen purchase. This technical feature realizes the operating logic of "prioritizing self-produced green hydrogen and supplementing with emergency external purchases," maximizing the utilization of green hydrogen produced from on-site renewable energy sources, reducing the frequency of using expensive externally purchased hydrogen, and significantly reducing the station's raw material procurement costs; the automatic switching mechanism based on inventory thresholds enables intelligent management and control of the external hydrogen purchase module, reducing human error; at the same time, the setting of a safe inventory lower limit prevents system shutdown caused by depletion of storage tank hydrogen, improves the system's risk resistance and operational self-sufficiency under extreme conditions, and balances the system's operational economy and hydrogen supply security.

[0044] In some specific embodiments, the mixed integer linear programming model is configured with asymmetric elastic mode switching constraints, wherein when the total available power output of wind and solar power is greater than the electrical load, it is an excess hydrogen production mode, and when the total available power output of wind and solar power is less than the electrical load, it is an under-discharge mode. When the system is in excess hydrogen production mode, constraints are applied. =0 forces the fuel cell to shut down; When the system is in under-discharge mode, apply constraints. =0 Forced shutdown of the electrolyzer, and no change to the media handling rate of the liquid hydrogen liquefaction unit Set a mandatory shutdown constraint.

[0045] = Hydrogen flow rate of fuel cell during time period t; =Power output of the electrolytic cell during time period t.

[0046] It should be understood that the system uses the relative values ​​of the total available wind and solar power output and the local electrical load as the threshold for judgment: when the total wind and solar power output is greater than the electrical load, it is in excess hydrogen production mode; when the total wind and solar power output is less than the electrical load, it is in under-discharge mode. In excess hydrogen production mode, through... =0 forces the fuel cell to shut down, avoiding unnecessary consumption of hydrogen storage; in under-discharge mode, through =0 forces the electrolyzer to shut down, stopping inefficient electrolysis operations, while not restricting the liquefaction unit's operating status. The liquefaction unit can continuously extract gaseous hydrogen from the high-pressure storage tank to complete the liquefaction process, and the fuel cell simultaneously consumes the stored hydrogen to generate electricity to make up for the power shortage. The entire mode switching logic breaks the traditional symmetrical operation mode of "synchronous start and stop of electrolyzer and liquefaction unit", realizing decoupled operation of the equipment.

[0047] For example, a pure photovoltaic off-grid power station operates the electrolyzer and shuts down the fuel cell when photovoltaic output is excessive during the day; at night, when photovoltaic output drops to zero and there is a significant power shortage, the electrolyzer is shut down, while the liquefaction unit continues to operate. Asymmetric elastic switching is the core design to ensure continuous liquid hydrogen production, completely solving the problem of synchronous shutdown of the liquefaction unit and insufficient liquid hydrogen production when there is insufficient power in traditional solutions. This design decouples the power supply from the hydrogen liquefaction process, allowing the liquefaction process to be unaffected by short-term power fluctuations and ensuring a stable daily output of liquid hydrogen. At the same time, it allows for precise start-up and shutdown of non-core equipment in different modes, reducing ineffective energy consumption and improving the utilization rate of renewable energy and hydrogen storage resources. This mode can perfectly adapt to the intermittent and fluctuating output characteristics of wind power and photovoltaic power, greatly improving the system's adaptability and production continuity under complex off-grid conditions.

[0048] In some specific embodiments, the mixed-integer linear programming model is further configured with equipment ramp-up rate constraints, specifically: Electrolytic cell power constraints: ; Hydrogen consumption constraints for fuel cells: ; Processing rate constraints of liquid hydrogen liquefaction units: ; in , , These are the upper limits of ramp rates for the electrolyzer, fuel cell, and liquid hydrogen liquefaction unit, respectively.

[0049] It should be understood that in engineering applications, the upper limit of the power ramp-up rate for electrolyzers is set at 20% / hour of rated power, the upper limit of the hydrogen consumption ramp-up rate for fuel cells is set at 15% / hour of rated hydrogen consumption, and the upper limit of the processing rate ramp-up rate for liquefaction units is set at 10% / hour of rated flow rate. Ramp-up constraints are used to limit the maximum variation in equipment power and hydrogen flow rate between two adjacent scheduling periods, preventing drastic changes in equipment load within a short period. During system operation, regardless of the switching of operating conditions, the load changes of the three types of equipment will be constrained within the corresponding threshold range, achieving a smooth load transition.

[0050] For example, when the system switches from a power surplus mode to a power shortage mode, the electrolyzer gradually reduces its load and shuts down, while the liquefaction unit's load is adjusted slowly, preventing sudden changes from full load to zero load. The equipment ramp-up rate constraint logically protects precision core equipment such as the electrolyzer, fuel cell, and cryogenic liquefaction unit. A smooth load transition significantly reduces internal mechanical and thermal stresses, decreasing the probability of equipment failure and effectively extending the overall service life of the equipment. This constraint also makes system condition switching smoother, avoiding safety hazards such as grid fluctuations and pipeline pressure surges caused by sudden load changes, thus improving the overall system's operational stability and safety. Simultaneously, the smooth load adjustment conforms to standard equipment operating specifications, reducing equipment maintenance frequency and further lowering the later-stage maintenance costs of the site, helping the system achieve long-term unattended and stable operation.

[0051] In some specific embodiments, the objective function expression is: The objective function includes ten cost items in sequence: wind power operation and maintenance cost, photovoltaic operation and maintenance cost, electrolyzer operation and maintenance cost, fuel cell operation and maintenance cost, hydrogen storage tank charging and discharging operation and maintenance cost, liquefaction reflux and re-storage cost, liquefaction unit operation and maintenance cost, wind and solar curtailment penalty cost, purchased hydrogen procurement cost, and electricity load unmet penalty cost.

[0052] = Corresponding equipment unit operation and maintenance cost coefficient; =Heat value of hydrogen (used for power-mass conversion); =Penalty coefficient for wind and solar power curtailment; = Curtailed wind / curtailed solar power; =Unit cost of purchased hydrogen; =Penalty factor for electrical load not meeting requirements; =Electric load deficit during period t; =Scheduling time step.

[0053] It should be understood that the objective function focuses on minimizing the total system operation and maintenance cost within a 24-hour scheduling cycle, with a time step of... The time frame is fixed at 1 hour, and T represents the total number of scheduling periods. The function contains ten costs in sequence: wind power operation and maintenance (O&M) costs, photovoltaic O&M costs, electrolyzer O&M costs, fuel cell O&M costs, hydrogen storage tank charging and discharging O&M costs, liquefaction reflux and re-storage costs, liquefaction unit O&M costs, wind and solar curtailment penalty costs, purchased hydrogen procurement costs, and unmet electricity load penalty costs. Each cost corresponds to a specific operational stage of the system, and the coefficients for each item represent the unit cost or penalty price of the corresponding unit. During model calculation, all cost items are comprehensively weighed, and under the premise of satisfying all constraints, the globally optimal scheduling scheme with the lowest total cost is found.

[0054] For example, under conditions of green electricity surplus, the model will appropriately reduce the amount of wind and solar power curtailment to decrease penalty costs; under conditions of power shortage, it will prioritize optimizing the balance between the costs of purchased hydrogen and load reduction. The complete full-chain cost objective function realizes the overall cost optimization of the off-grid liquid hydrogen production system from power generation, hydrogen production, hydrogen storage, liquefaction to electricity load, moving away from the one-sidedness of traditional solutions that only calculate the cost of a single link; the ten cost items are clearly designed, making it easy for technical personnel to break down and analyze the cost ratio of each link, identify high-energy-consuming and high-cost units, and carry out targeted energy-saving and cost-reduction optimization; the scheduling scheme based on this objective function has the economic characteristics of global optimization, rather than local optimization, and can continuously reduce the overall operating cost of the station; at the same time, the setting of penalty cost items also guides the system to prioritize the consumption of renewable energy and ensure electricity load, taking into account both economic efficiency and energy utilization efficiency.

[0055] In some specific embodiments, the electrical load in the objective function does not meet the penalty coefficient. The value is greater than the unit price of purchased hydrogen. ; In the model's execution logic, the mandatory hard constraint corresponding to the daily liquid hydrogen production has the highest priority. When there is a shortage of green electricity, the system prioritizes making up for the power shortage by purchasing hydrogen from outside, and then reduces the local power load. Throughout the process, the system prioritizes ensuring that the daily liquid hydrogen production target is achieved.

[0056] It should be understood that the electrical load did not meet the penalty factor. The value is much greater than the unit price of purchased hydrogen. This numerical gap serves as a crucial threshold for model decision-making; simultaneously, the daily liquid hydrogen production constraint is explicitly defined as the highest priority constraint, and all cost optimization and load adjustment actions must be carried out under the premise of achieving the liquid hydrogen production target. When the system encounters a green electricity shortage, the model makes decisions according to a fixed priority order: first, ensuring that the daily liquid hydrogen production target is met; second, prioritizing the purchase of hydrogen to supplement the electricity gap; and only when the purchase of hydrogen still cannot make up for the gap, selectively reducing local power load. The system operates entirely according to this priority logic, and will not sacrifice core production targets in pursuit of low costs.

[0057] For example, under extreme operating conditions with severe shortages of green electricity, the system will not drastically reduce liquefaction capacity to avoid the cost of purchasing hydrogen from external sources. Instead, it will start purchasing hydrogen from external sources to maintain production, only slightly reducing the electricity load. This coefficient setting and priority logic firmly safeguard the core production target of daily liquid hydrogen output from a rule-based perspective, completely eliminating the problem of the traditional solution's "emphasis on electricity consumption and neglect of hydrogen production" inversion of priorities. The reasonable coefficient difference allows the system to prioritize purchasing hydrogen from external sources rather than directly cutting off the load, maximizing the normal supply of local electricity load and balancing production demand and electricity demand. The entire decision-making logic is intelligent and standardized, making optimal choices in the face of various extreme operating conditions, greatly improving the system's fault tolerance and adaptability. At the same time, the fixed priority framework also makes the scheduling logic predictable and controllable, facilitating the site to formulate production plans and emergency plans, and improving the overall operation and management level.

[0058] Second Embodiment A hierarchical elastic constraint optimization method for an off-grid liquid hydrogen production system according to an embodiment of this application includes the following steps: Collect the available wind power output sequence, available photovoltaic power output sequence, and local electricity load demand sequence for each time period within the scheduling cycle; The total available output of wind power and photovoltaic power is compared with the corresponding electricity load for each time period to predict whether the current operating mode is excess hydrogen production mode or insufficient discharge mode. A mixed-integer linear programming model is constructed with the optimization objective of minimizing the total operation and maintenance cost of the scheduling cycle. In the model, the liquid hydrogen storage tank's inventory at the end of the scheduling cycle is equal to the preset daily liquid hydrogen production target, which is set as the first-level mandatory hard constraint. An unmet electrical load variable is introduced into the electrical balance equation, and the local electrical load is set as the second-level optional soft constraint. A penalty cost term is added to the unmet electrical load in the model objective function. Solve the constructed mixed-integer linear programming model to output the operating status, power allocation, and media flow scheduling scheme of all devices in each time period within the scheduling cycle.

[0059] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system, characterized in that, include: The data acquisition module is used to collect the available wind power output sequence, available photovoltaic power output sequence, and local electricity load demand sequence for each time period within the scheduling cycle. The optimized control module is used to compare the total available output of wind power and photovoltaic power with the corresponding electrical load on a time-by-time basis, and to predict whether the current operating mode is excess hydrogen production mode or insufficient discharge mode. The configuration module is used to construct a mixed integer linear programming model with the optimization objective of minimizing the total operation and maintenance cost of the scheduling cycle. In the model, the liquid hydrogen storage tank's inventory at the end of the scheduling cycle is equal to the preset daily liquid hydrogen production target, which is set as the first-level mandatory hard constraint. The unmet electrical load variable is introduced into the electrical balance equation, and the local electrical load is set as the second-level optional soft constraint. A penalty cost term is added to the unmet electrical load in the model objective function. The output module is used to solve the constructed mixed-integer linear programming model and output the operating status, power allocation and media flow scheduling scheme of all devices in each time period within the scheduling cycle.

2. The hierarchical elastic constraint optimization system of an off-grid liquid hydrogen production system according to claim 1, wherein, The mixed-integer linear programming model includes a dynamic equation for the high-pressure hydrogen storage tank inventory: ； The equation includes a liquefied reflux term The liquefied reflux term represents the mass flow rate of gaseous hydrogen that is not liquefied and is refluxed to the high-pressure hydrogen storage tank from the liquid hydrogen liquefaction device.

3. The hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system according to claim 2, characterized in that, The liquidized backflow re-storage operation and maintenance cost term is added in the objective function of the mixed integer linear programming model .

4. The hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system according to claim 3, characterized in that, The mixed-integer linear programming model is configured with mutual exclusion constraints for hydrogen charging and discharging and minimum start-up constraints for the liquefaction unit, and introduces binary variables of the hydrogen charging state. Binary variable of hydrogen release state and impose constraints ; Introducing binary variables for liquefaction device switches and impose constraints The liquefaction unit is limited to operating only within the range of rated minimum to maximum output.

5. The hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system according to claim 4, characterized in that, The mixed-integer linear programming model is also configured with coupling constraints based on the condition of purchased hydrogen, and introduces auxiliary binary variables. ; When the high-pressure hydrogen storage tank level is higher than the safety lower limit, forced ; When the high-pressure hydrogen storage tank is below the safety limit, the purchase of hydrogen from external sources is permitted.

6. The hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system according to claim 5, characterized in that, The mixed-integer linear programming model is configured with asymmetric elastic mode switching constraints, wherein when the total available power of wind and solar power is greater than the electric load, it is in the excess hydrogen production mode, and when the total available power of wind and solar power is less than the electric load, it is in the under-discharge mode. When the system is in excess hydrogen production mode, constraints are applied. =0 forces the fuel cell to shut down; When the system is in under-discharge mode, apply constraints. =0 Forced shutdown of the electrolyzer, and no change to the media handling rate of the liquid hydrogen liquefaction unit Set a mandatory shutdown constraint.

7. The hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system according to claim 6, characterized in that, The mixed-integer linear programming model is also configured with equipment ramp-up rate constraints, specifically: Electrolytic cell power constraints: ; Hydrogen consumption constraints for fuel cells: ; Processing rate constraints of liquid hydrogen liquefaction units: ; in , , These are the upper limits of ramp rates for the electrolyzer, fuel cell, and liquid hydrogen liquefaction unit, respectively.

8. The hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system according to claim 7, characterized in that, The objective function expression is: The objective function includes ten cost items in sequence: wind power operation and maintenance cost, photovoltaic operation and maintenance cost, electrolyzer operation and maintenance cost, fuel cell operation and maintenance cost, hydrogen storage tank charging and discharging operation and maintenance cost, liquefaction reflux and re-storage cost, liquefaction unit operation and maintenance cost, wind and solar curtailment penalty cost, purchased hydrogen procurement cost, and electricity load unmet penalty cost.

9. The hierarchical elastic constraint optimization system for an off-grid liquid hydrogen production system according to claim 8, characterized in that, The objective function includes a penalty coefficient for electrical load not meeting the requirement. The value is greater than the unit price of purchased hydrogen. ; In the model's execution logic, the mandatory hard constraint corresponding to the daily liquid hydrogen production has the highest priority. When there is a shortage of green electricity, the system prioritizes making up for the power shortage by purchasing hydrogen from outside, and then reduces the local power load. Throughout the process, the system prioritizes ensuring that the daily liquid hydrogen production target is achieved.

10. A hierarchical elastic constraint optimization method for an off-grid liquid hydrogen production system, characterized in that, A hierarchical elastic constraint optimization system applied to an off-grid liquid hydrogen production system according to any one of claims 1 to 9, comprising the following steps: Collect the available wind power output sequence, available photovoltaic power output sequence, and local electricity load demand sequence for each time period within the scheduling cycle; The total available output of wind power and photovoltaic power is compared with the corresponding electrical load for each time period to predict whether the current operating mode is excess hydrogen production mode or insufficient discharge mode. A mixed-integer linear programming model is constructed with the optimization objective of minimizing the total operation and maintenance cost of the scheduling cycle. In the model, the liquid hydrogen storage tank's inventory at the end of the scheduling cycle is equal to the preset daily liquid hydrogen production target, which is set as the first-level mandatory hard constraint. An unmet electrical load variable is introduced into the electrical balance equation, and the local electrical load is set as the second-level optional soft constraint. A penalty cost term is added to the unmet electrical load in the model objective function. Solve the constructed mixed-integer linear programming model to output the operating status, power allocation, and media flow scheduling scheme of all devices in each time period within the scheduling cycle.

Images