Hydrogen storage energy optimization configuration method for integrated energy system based on health state perception and whole life cycle optimization

Abstract

A comprehensive energy system hydrogen storage optimization configuration method based on health status perception and full life cycle optimization includes: establishing a fuel cell health status perception model using an exponential decay model; formulating multi-module collaborative operation strategies to address the dynamic differences in module health; constructing an energy conversion model for hydrogen storage units using a unified energy index; establishing a full life cycle optimization configuration model based on the adaptability of long-term cost and short-term operation; constructing a full life cycle health perception multi-energy optimization model by integrating the coupling requirements of electricity, heat, and gas; and solving the full life cycle health perception multi-energy optimization model by inputting the time-series data of the park's source and load and equipment parameters to obtain the number of modules, capacity, and load allocation strategy. This method achieves coordinated optimization of the economy, reliability, and low-carbon characteristics of hydrogen storage configuration in multi-energy coupling systems by constructing a dynamic health status model of equipment, designing multi-module collaborative operation strategies, and integrating full life cycle cost accounting.

Description

Technical Field

[0001] This invention belongs to the field of hydrogen energy storage optimization configuration technology for integrated energy systems, specifically involving a method for optimizing the configuration of hydrogen energy storage in integrated energy systems based on health status perception and full life cycle optimization. Background Technology

[0002] In existing technologies, industrial parks, as core scenarios for energy consumption and carbon emissions, are accelerating their transformation into integrated energy systems that couple electricity, heat, and gas. With the large-scale grid connection of renewable energy sources such as wind and solar power, the intermittency and volatility of their output exacerbate the spatial and temporal mismatch between energy supply and demand, leading to prominent challenges in renewable energy integration. Hydrogen energy storage, with its high energy density, long-term storage, and multi-energy conversion characteristics, has become a key supporting technology for mitigating the fluctuations in renewable energy and ensuring stable system operation, and has received widespread attention and application in integrated energy systems within industrial parks.

[0003] However, existing optimization configuration and operation scheduling methods for hydrogen energy storage systems still have significant limitations: First, they generally ignore the dynamic health evolution of fuel cells, treating the equipment as an ideal operating state and failing to consider performance degradation caused by cumulative load. This leads to premature failure of some modules in multi-module systems due to overuse, shortening the overall lifespan. Second, the equipment's total lifecycle cost and short-term operational optimization are disconnected. Traditional configuration schemes often focus on initial investment costs, failing to incorporate long-term costs such as operation and maintenance, health degradation calculations, and decommissioning into the optimization framework, resulting in poor long-term economic viability. Third, multi-module collaborative operation strategies lack a health-state-oriented dynamic adjustment mechanism, making it difficult to balance the load distribution and lifespan equilibrium of each module. This can easily lead to extreme differentiation in module health states, affecting system reliability. Fourth, in multi-energy coupling scenarios involving electricity, heat, and gas, existing multi-energy flow optimization models do not fully integrate equipment health state constraints, making it difficult to balance energy conversion efficiency and equipment lifespan assurance, thus hindering the improvement of overall system benefits. Although the hydrogen energy storage industry has formed a complete "production-storage-transportation-application" chain, insufficient equipment lifecycle management capabilities remain a pain point for the industry. As the core conversion device in hydrogen energy storage systems, the lifespan and operating efficiency of hydrogen fuel cells directly determine the economic viability of projects. However, in actual operation, factors such as cumulative load and fluctuations in operating conditions can lead to a non-linear decline in their health status. Existing technologies lack a deep correlation between health status and configuration schemes and operating strategies, employing fixed-parameter models for system design. This not only results in significant deviations between optimized results and actual operation but may also drive up the total lifecycle cost due to premature equipment failure. Furthermore, the lack of coordinated consideration between the energy balance constraints of multi-energy coupled systems and the health constraints of hydrogen energy storage devices further exacerbates the complexity of system scheduling and operational risks.

[0004] Therefore, there is an urgent need for a method to optimize hydrogen energy storage configuration that can accurately sense the health status of fuel cells and take into account the entire life cycle cost. Summary of the Invention

[0005] To address the issues of imbalanced lifecycle costs, neglected fuel cell health status, and insufficient benefits of multi-module collaboration and multi-energy coupling in hydrogen energy storage configurations within integrated energy systems, this invention proposes a method for optimizing hydrogen energy storage configuration in integrated energy systems based on health status perception and lifecycle optimization. This method achieves synergistic optimization of the economy, reliability, and low-carbon aspects of hydrogen energy storage configuration in multi-energy coupling systems by constructing a dynamic model of equipment health status, designing multi-module collaborative operation strategies, and integrating lifecycle cost accounting. This provides technical support for the long-term stable operation of integrated energy systems in industrial parks and promotes the large-scale commercial application of hydrogen energy storage technology.

[0006] The technical solution adopted in this invention is as follows:

[0007] The integrated energy system hydrogen storage optimization configuration method based on health status perception and full life cycle optimization includes the following steps: Step 1: Establish a fuel cell health status perception model using an exponential decay method; Step 2: Based on the fuel cell health status perception model built in Step 1, formulate a multi-module collaborative operation strategy to address the dynamic differences in module health. Step 3: Construct an energy conversion model for the hydrogen energy storage unit using a unified energy index; Step 4: Based on the adaptability between long-cycle costs and short-term operation, establish a full lifecycle optimization configuration model; Step 5: Based on the hydrogen energy storage unit energy conversion model in Step 3 and the full life cycle optimization configuration model in Step 4, integrate the multi-energy coupling requirements of electricity, heat and gas to construct a full life cycle health perception multi-energy optimization model; Step 6: Using the MATLAB simulation environment and CPLEX optimization solver, input the source-load time series data and equipment parameters of the park, solve the full life cycle health sensing multi-energy optimization model of Step 5, and obtain the number of modules, capacity and load allocation strategy.

[0008] Step 7: Set up multiple comparison schemes and compare and analyze the annualized cost, carbon emission reduction rate, renewable energy consumption rate and module SOH evolution trajectory corresponding to the solution results of Step 6 under typical day scenarios in the four seasons to verify the effectiveness of the proposed method.

[0009] In step 1, based on the physical degradation law of fuel cells and combined with the impact of cumulative load on lifespan, a fuel cell health status perception model is established using an exponential decay method. The specific construction method is as follows: (1); In formula (1): This represents the percentage of the fuel cell module i in terms of its health status; L represents the cumulative output power. This represents the maximum load during the planning period. Specific attenuation coefficient for the module; Indicates the total number of fuel cell modules; This is the exponential decay control parameter, typically A value greater than 1 describes the nonlinear amplification effect of load intensity on health decline, meaning that the life loss caused by bearing a high load in a short period of time is much higher than the total loss caused by bearing a low load over a long period of time.

[0010] This model can reflect the nonlinear characteristics of rapid degradation of modules under high load conditions and extended lifespan of modules under low load conditions, providing a basis for evaluating the feasibility of different module combination schemes during the configuration phase.

[0011] Furthermore, the degradation process can be described using the module lifetime consumption rate formula, as shown below: (2); In formula (2): This indicates that the i-th fuel cell module is under the cumulative equivalent load The health decay rate under certain conditions is used to describe the trend of the module's state of health (SOH) as the operating load accumulates; Represents the cumulative load L The power is used to describe the nonlinear change in degradation rate with increasing operating load.

[0012] The above formula quantifies the instantaneous lifespan consumption rate of modules under different loads, which can intuitively reflect that high-load modules degrade faster than low-load modules, providing a basis for load distribution and multi-module collaborative configuration.

[0013] In a multi-module system, a weighted average indicator can be introduced to evaluate the overall health status, as shown below: (3); In formula (3): This represents the weighted average of the overall health status of a multi-module fuel cell system, with a value range of [0,1]. The module weight can be determined based on the module's importance or expected load allocation. This indicator reflects the overall health level of the fuel cell stack.

[0014] Furthermore, to ensure that the configuration scheme can meet future load demands in the long term, it is necessary to consider the constraints of the current remaining lifespan on the carrying capacity of future loads: (4); In equation (4): This represents the equivalent cumulative load that the module will need to bear in the future during the planning period; Indicates the current remaining lifespan percentage; This indicates that the current accumulated load of the i-th fuel cell module is... The proportion of people in a healthy state at that time; This indicates the current accumulated operating load of the module; this constraint incorporates the matching of health status with future load into the configuration evaluation, and is a prerequisite for achieving full lifecycle optimization.

[0015] In step 2, based on the fuel cell health status perception model constructed in step 1, dynamic power allocation and lifetime balancing are incorporated to formulate a multi-module collaborative operation strategy, taking into account the dynamic differences in module health. The specific construction method is as follows: To prevent modules with low SOH from continuing to withstand high loads and thus accelerating degradation, the module i During the period t The allowed operating range constraints are as follows: (5); In formula (5): Representation module i During the period t The actual output power is the optimization decision variable, representing how much load the system ultimately decides to let the module bear at that moment; Representation module i The rated maximum output power that can be provided when SOH=1; Representation module i During the period t The health status, usually .

[0016] Equation (5) reflects the basic relationship that the higher the health level, the greater the power that can be carried, and the lower the health level, the more the upper limit of output automatically shrinks. This is the basis for the health status to participate in the operation decision.

[0017] Secondly, in order to dynamically track the degradation process of each module in the optimization model, it is necessary to construct an evolution model of SOH with operating load. Based on the high load-high degradation characteristics of fuel cells, this section adopts the following load intensity-driven degradation expression: (6); In formula (6): Representation module i In the next period t+ 1. Health status; Representation module i During the period t ; output power; Representation module i Rated power; Indicates the coefficient of health decline; Indicates the optimization time step; The amplification effect of high load on the sensitivity to decay is described; Equation (6) is a discretized approximation of Equation (1) and is used for annual time-series operation simulation. By expressing the cumulative load as the power integral over time, it can be proven that the two models are consistent in the long-term trend at small time steps. Equation (6) and Equation (5) above form a two-way closed loop where output affects health and health limits output, enabling the model to naturally inhibit the continuous use of modules with weak health.

[0018] However, simply allowing each module to run independently within its health range is insufficient to achieve lifetime-friendly behavior. To prevent a highly uneven lifetime distribution caused by some modules running under heavy loads for extended periods while others remain idle, a health deviation constraint between modules is introduced. This constraint guides optimization to tilt the load towards modules with higher health levels, while preventing the health deviation from widening further. The health deviation constraint between modules can be expressed as: (7); In equation (7): Representation module i With modules j During the period t Health deviations; This represents the health status ratio of the i-th fuel cell module in time period t, with a value range of [0,1]. It is a real-time quantitative indicator of the health status of a single module. This represents the percentage of the j-th fuel cell module in good health during time period t, and is related to... The definition is consistent and is only used to distinguish any other module in a multi-module system, ensuring that health differences are covered across all module combinations. This represents the maximum permissible health difference threshold, typically set to 0.05~0.15; Represents the discretized scheduling time; This indicates the total number of time periods within the planning period, used to limit the time coverage of health difference constraints and ensure the balanced health status of modules throughout the entire operation cycle.

[0019] Within the framework of lifecycle optimization, this invention primarily relies on the health degradation cost in the objective function. This approach flexibly guides the balancing of lifespan among modules. The constraint on health differences between modules serves as an auxiliary boundary condition, preventing extreme divergence in health states within a multi-module system. It ensures that health differences between modules remain within a reasonable range permissible for safe system operation, thereby guaranteeing lifespan balancing and sustainability during group collaborative operation.

[0020] Building upon this, to ensure the model not only focuses on the module's availability in the current period but also assesses its continued service capability over future operating cycles, a future availability capacity prediction based on health status is further introduced. Given that SOH essentially reflects the module's remaining lifetime, its equivalent usable capacity can be inferred based on future load levels, thereby achieving a quantitative assessment of the module's lifetime sustainability, as detailed below: (8); In equation (8): Representation module i During the period t The remaining load capacity is quantified to determine its sustainable operating capacity over a future period. Through this model, the system can anticipate the future load capacity of each module during the planning stage, thereby achieving a balance between forward-looking load allocation and health protection.

[0021] In step 3, combining the bidirectional electro-hydrogen-thermal conversion characteristics and based on electrochemical and thermochemical principles, a unified energy index is used to construct an energy conversion model for the hydrogen energy storage unit; the specific construction method is as follows: (1) Electrochemical characteristics: In a hydrogen energy storage unit, the electrolyzer and fuel cell serve the functions of consuming electricity to produce hydrogen and burning hydrogen to generate electricity, respectively. Their hydrogen-to-electricity conversion models should be shown in equations (9) and (10), respectively: (9); (10); In the above formula: , They are respectively t Power consumption and hydrogen production capacity of the time-phase electrolyzer; , They are respectively t The power consumption and hydrogen production capacity of the fuel cell during different time periods; and These represent the energy conversion efficiencies for hydrogen production using electricity consumed in an electrolyzer and for power generation using hydrogen consumed in a fuel cell, respectively. For simplicity, the energy conversion efficiency is typically set as a constant in hydrogen energy storage unit configuration studies.

[0022] To facilitate system modeling and energy flow analysis, this invention employs a unified energy index to characterize the dynamic conversion process of electrical energy, thermal energy, and hydrogen energy in a hydrogen energy storage unit. Based on the principle of energy equivalence conversion, the hydrogen storage capacity can be converted into standard energy units through its calorific value parameter. Specifically, under standard conditions (0°C, 1 atmosphere), the lower heating value of hydrogen per unit volume is taken as 2.95 kW·h / m³, thereby achieving an equivalent conversion between hydrogen physical quantities (volume / mass) and energy values. This standardized approach not only ensures the consistency of energy measurement in multi-energy flow coupled systems but also provides a unified quantitative benchmark for system optimization.

[0023] (2) Thermochemical properties: The thermodynamic properties of electrolyzers and fuel cells can be characterized by the following equations: (11); (12); In the formula: The operating temperature of the electrolyzer during time period t is the electrolyte temperature. The heat capacity of the electrolytic cell; The operating temperature of the fuel cell during time period t; The heat capacity of the fuel cell; , These represent the heat absorption power of the cooling circulating water in the electrolyzer and fuel cell during time period t, respectively. Let t be the power consumption of the electrolytic cell during time period t; The hydrogen production power of the electrolyzer during time period t; The hydrogen consumption power of the electrolyzer during time period t; Let t be the power consumption of the electrolytic cell during time period t.

[0024] In step 4, the fuel cell health status perception model from step 1 is integrated with the multi-module collaborative operation strategy from step 2. Based on the adaptability between long-term costs and short-term operation, a health degradation cost discounting mechanism is incorporated to establish a full life-cycle optimization configuration model. The specific construction method is as follows: Since direct time-series optimization of the total lifecycle cost needs to cover the entire service life of the equipment, which typically lasts 20 to 30 years, and the source-load data of the integrated energy system in industrial parks has significant annual repetition characteristics, directly conducting long-term hourly simulations would lead to an exponential increase in computational complexity. Therefore, this invention is based on the Equivalent Annual Cost (EAC) method, first constructing a cost quantification model in the entire lifecycle dimension, and then converting it into an annualized total lifecycle cost for subsequent modeling and solving; wherein, the long-term replacement cost caused by fuel cell health degradation is allocated to each operating year to form a corresponding annualized health degradation cost; The objective function for the total lifecycle cost of the system is as follows: (13); In equation (13): The total cost of a fuel cell subsystem throughout its entire lifecycle is a core indicator for measuring the long-term economic viability of the system. N This represents the total number of fuel cell modules; For module i One-time investment cost; , , and Modules i Future load during the planning period The calculated operating and maintenance costs, calculated carbon emission costs, calculated operating penalty costs, and calculated energy purchase costs are as follows; specifically: This indicates that module i, during the planning period, is responsible for assuming the equivalent cumulative load of the future. The resulting operation and maintenance costs, spread over the entire life cycle, include costs for routine maintenance, replacement of consumables (such as fuel cell catalysts and electrolytes), and equipment inspections. Indicates that module i is responsible for The total life-cycle environmental cost, calculated based on the carbon emission price during the planning period, for carbon emissions generated indirectly or directly, includes carbon emissions from purchased energy sources (such as grid electricity and natural gas) associated with module operation, as well as implicit carbon emissions from equipment manufacturing / replacement. This indicates that module i is affected by unreasonable load distribution (e.g.) (Exceeding the health carrying capacity, resulting in insufficient output), causing problems such as power outage / heat loss, wind curtailment, and solar curtailment, and the full life cycle economic loss calculated according to the planning period penalty rules; Indicates that module i is responsible for The energy purchase cost (electricity, natural gas) incurred from purchasing energy (electricity, natural gas) from external power grids / gas grids, calculated over the entire life cycle, is the core variable cost of module operation.

[0025] Representation module i Future period t The equivalent load that needs to be borne.

[0026] The full life cycle cost of the fuel cell subsystem defined in the above formula provides a basis for calculating the annualized cost model for subsequent expansion to the park's integrated energy system. Its core cost items will be converted into annual costs using the equal annual value method, and then optimized and solved after being integrated with the system-level expansion cost items.

[0027] To incorporate health status into configuration optimization, the constraints on system capacity from the future equivalent load requirements of modules are as follows: (14); In equation (14): For module i Rated output capacity; To configure the initial module health status ratio; function

[0028] This is used to quantify the current health status and its limitations on future load capacity, ensuring that the system will not experience insufficient load due to module aging during the planning period.

[0029] The choice of module capacity and quantity directly affects future load capacity and cost consumption. Therefore, capacity constraints are introduced as follows: (15); In equation (15): To meet the maximum load demand during the planning period, the configuration scheme must be able to meet the load at any time.

[0030] To comprehensively consider lifespan depletion and economic efficiency, a health degradation cost is further defined. This cost transforms the module's lifespan loss into quantifiable economic depreciation and is explicitly incorporated into the objective function. The calculation formula is shown below: (16); In equation (16): It is expressed as the total cost of health degradation over the entire life cycle, which is the total replacement cost corresponding to the life loss of the equipment throughout its entire service life. This represents the cost of replacing the i-th fuel cell module once. and These are the module's health status at the beginning and end of the planning period, respectively; This is the health threshold for the end of life, typically 0.7; The discount factor is given by equation (17). (17); In equation (17): It represents the annual discount rate, also known as the time value of money, which reflects the ability of current funds to increase in value relative to future funds. The time period (in years) representing the period from the beginning of the planning period to the actual occurrence of the cost is a key variable for determining the discount rate. for tThe discount factor for a given period is used to convert future replacement costs into their present value. This lifespan discounting method assumes that the performance degradation of a fuel cell is approximately linearly related to its remaining lifespan within its effective lifespan. This assumption has been adopted by numerous engineering practices and lifespan assessment studies and is suitable for long-term configuration and economic analysis scenarios.

[0031] The subsequent steps will be based on formula (16), i.e. According to the service life of the equipment m With interest rates r Convert it into an annualized cost of health degradation. This is incorporated into the annual optimization goals, thus linking the lifecycle loss with the annual operating strategy. The final configuration optimization problem can be formalized as: (18); In formula (18): The set of decision variables for module capacity; N To optimize the number of modules in the object; This represents the minimum capacity of the i-th module; This represents the maximum capacity of the i-th module; This represents the minimum number of fuel cell modules; This represents the maximum number of fuel cell modules; in this model The total lifecycle cost of the fuel cell subsystem is characterized. To adapt to the annual time-series simulation requirements of the integrated electric-heat-gas energy system, subsequent research will convert it into an annualized total lifecycle cost using the equal-annual value method. It is also integrated with system-level extended cost items such as equipment investment and purchased energy in the electric-heat-gas integrated energy system, realizing the extension and integration of the subsystem cost model into the system-level optimization model.

[0032] In step 5, based on the hydrogen energy storage unit energy conversion model in step 3 and the full life cycle optimization configuration model in step 4, and integrating the multi-energy coupling requirements of electricity, heat, and gas, a full life cycle health sensing multi-energy optimization model is constructed, setting constraints such as energy balance, pipeline safety, and hydrogen blending ratio. The specific construction method is as follows: Based on total life cycle cost Total cost of health degradation To balance the optimization objectives of the entire lifecycle with the solution requirements of annual time-series simulations, this section transforms the costs into annualized comprehensive lifecycle costs using the equal-annual value method. This is then used as the objective to achieve coordinated optimization of the hydrogen energy storage system's configuration and operation. Specifically, each cost is transformed into an annual cost using the following two annualization methods: ①: For one-time costs such as initial investment and total life-cycle health degradation costs: the equivalent annual value method is used, i.e., through the capital recovery factor. Converted to annual equivalent costs, of which, Indicates the service life of the equipment; ②: For operating costs incurred in different time periods, such as energy purchase costs, carbon emission costs, penalty costs, and operation and maintenance costs: directly take the sum of their annual incurrence values, which is the reflection of the time-period cost of the entire life cycle in a single year; as detailed below: The core characteristic of time-of-use operating costs (energy purchase, carbon emissions, penalties, and operation and maintenance costs) is that they occur dynamically over time and repeat annually. Since the source-load data (wind / solar power output, electricity / heat load) of industrial parks exhibits significant annual repetition, the time-of-use cost structure is highly similar year by year throughout the entire lifecycle, eliminating the need for simulating 20-30 year long-cycle data annually. Therefore, directly taking the sum of time-of-use costs for a single year can effectively represent the average annual level of this type of cost throughout the entire lifecycle, simplifying computational complexity without affecting the accuracy of the optimization results, and is compatible with the equivalent annual value method.

[0033] Specifically, as shown in equations (21), (22), (23), and (24), the general formula can be expressed as: ; in: This represents the annual aggregate value of a certain type of operating cost across different time periods, such as annualized energy purchase cost, annualized carbon emission cost, etc. This indicates the total number of time periods throughout the year, ensuring coverage of the entire annual operating cycle; This represents the unit cost coefficient for time period t, such as electricity price, carbon emission converted unit price, penalty unit price, unit power operation and maintenance cost, etc. This represents the cost-related physical quantities during time period t, such as purchased electricity capacity, purchased gas volume, power outage load, and equipment operating power. This indicates the optimization time step, with a default value of 1 hour.

[0034] (1) Objective function: Of the two methods: Equations (20) and (25) are the first method; Equations (21), (22), (23), and (24) are the second method; both methods correspond to the full lifecycle cost in step 4, ensuring the compatibility between long-term optimization objectives and short-term time-series simulations, as shown in Equation (19): (19); (20); (twenty one); (twenty two); (twenty three); (twenty four); (25); In the above formula, This represents the annualized total lifecycle cost of hydrogen energy storage configuration in an integrated energy system combining electricity, heat, and gas. A smaller value indicates better long-term economic performance of the system. Essentially, it uses the equal-annual cost method to convert one-time costs (such as initial investment and health degradation costs) and time-period operating costs (such as energy purchase and maintenance costs) into annual equivalent costs, thus achieving a match between long-term costs and short-term time-series simulations. , , , , and These represent annualized investment cost, annualized energy purchase cost, annualized carbon emission conversion cost, annualized operating penalty cost, annualized operation and maintenance cost, and annualized health degradation cost, respectively.

[0035] Among them: annualized investment cost Compared with the unit capacity investment cost of electrolyzers, fuel cells, and hydrogen storage tanks , , and configuration capacity , , Related; Annualized energy purchase cost Depend on t Time-of-use electricity pricing Gas prices With power purchase Gas purchase volume rate Combined with unit duration (This invention uses 1 hour) and the number of time periods throughout the year. calculate; Annualized carbon emission cost of , These are the carbon emission costs per unit of electricity and per unit volume of natural gas, respectively, obtained by multiplying the carbon emission price by the carbon emission factor. Annualized operating penalty cost Involving t Periodic power outage load Heat loss load Wind and solar power curtailment And the unit price for power outages and heat loss load penalties , and wind and solar curtailment penalty coefficient ; Annualized operating and maintenance costs With electrolyzers, fuel cells and hydrogen storage tanks tTime-based unit power operation and maintenance cost , , and operating power , , related.

[0036] In this model Minimizing is equivalent to total lifecycle cost The minimization of the two, and their quantification relationship is shown in equation (26), that is... yes Equivalent decomposition at the annual level; (26); In the above formula, The total lifecycle cost of a fuel cell subsystem (unit: yuan) is the sum of fixed and variable costs throughout the entire lifecycle of the equipment, from commissioning to decommissioning. This represents the total cost of health degradation over the entire lifecycle of the fuel cell subsystem (unit: yuan), which is the total replacement cost incurred due to the loss of lifespan caused by performance degradation (decline in state of health, SOH).

[0037] (2) Constraints: 1) Power balance constraints: (27); In equation (27): , They are respectively t Solar and wind power output during different time periods; for t Electrical load during a given time period; for t The power generation capacity of the gas turbine during a given period; This represents the amount of wind and solar power curtailed during time period t; This represents the amount of electricity purchased from the external power grid during time period t; This represents the power loss of the load during time period t; This represents the power consumption of the electrolytic cell during time period t; This represents the power generation of the fuel cell during time period t.

[0038] 2) Thermal energy balance constraint: In the IES's thermal energy supply architecture, the local heat load is shared by the gas turbine unit and the hydrogen energy system, which satisfies: (28); In equation (28): , They are respectively t Heat production capacity of gas turbines and gas boilers during specific time periods; fort Heat load during a given period; This represents the heat output power of the fuel cell during time period t. This represents the heat generation power of the electrolytic cell during time period t; This represents the heat loss load power during time period t.

[0039] 3) Gas energy balance constraint: As shown in equation (29), the gas supply for Class 2 gas-fired equipment is provided by purchased natural gas and hydrogen mixed into the pipeline network.

[0040] (29); In equation (29): This represents the rate at which gas is purchased from the external gas network during time period t. This refers to the calorific value of natural gas (in kW·h). , They are respectively t Gas consumption of gas turbines and gas boilers during a given period (converted to power based on the calorific value of natural gas); The hydrogen mixing power of the natural gas pipeline network during time period t 4) Operating constraints of gas equipment: The electro-thermal energy conversion relationship between gas turbines and gas boilers, as well as the upper and lower limits of their output, are shown in equations (30) to (349): (30); (31); (32); (33); (34); In the above formula: , These are the power generation efficiency and waste heat utilization efficiency of the gas turbine, respectively. The heat production efficiency of a gas-fired boiler; , These are the maximum power generation capacity of the gas turbine and the maximum heat production capacity of the gas boiler, respectively. This represents the power generation capacity of the gas turbine during time period t; This represents the gas consumption of the gas turbine during time period t; This represents the gas consumption of the gas-fired boiler during time period t.

[0041] 5) Operational constraints of hydrogen energy storage units: The upper and lower limits of power for electrolyzers and fuel cells are shown in Equations (35) and (36), respectively, and the upper and lower limits of SOC for hydrogen storage and SOC for thermal storage are shown in Equations (37) and (38), respectively.

[0042] (35); (36); (37); (38); In the above formula: This represents the power consumption of the electrolytic cell during time period t; Indicates the rated capacity of the electrolytic cell; This represents the power generation of the fuel cell during time period t; Indicates the rated capacity of the fuel cell; This represents the state of charge (SOC) of hydrogen storage during time period t. Let t be the state of charge (SOC) of hydrogen storage. Let SOC be the thermal storage state at time t; , and , These are the minimum and maximum values ​​of the hydrogen storage SOC and thermal storage SOC, respectively.

[0043] During the configuration process, it is generally necessary to ensure that the SOC value remains unchanged at the beginning and end of the daily scheduling cycle, as shown in equations (39) and (40): (39); (40); In the formula: T The number of time periods in a daily scheduling cycle; This represents the initial value of the hydrogen storage SOC during the daily scheduling cycle; This represents the final state value of hydrogen storage SOC during the daily scheduling cycle; This is the initial value of the thermal storage SOC during the daily scheduling cycle; This represents the final state value of the thermal storage SOC during the daily scheduling cycle; 6) System piping constraints: Considering the physical constraints of the actual energy network, the amount of electricity purchased from the external power grid and the amount of gas purchased from the external gas grid should both be kept within a certain range, that is, they must meet the following requirements: (41); (42); In the above formula: The power purchased from the external power grid; The gas purchase volume rate of the external gas network; , These are the maximum power purchase capacity of the external power grid and the maximum gas purchase volume rate of the external gas grid, respectively.

[0044] In addition, considering the safety of the natural gas pipeline network and the combustion performance of terminal appliances, the volume of hydrogen mixed in natural gas should be maintained within a certain range, that is, it must meet the following requirements: (43); In equation (43): This refers to the calorific value of hydrogen. This represents the maximum hydrogen mixing ratio; This represents the hydrogen mixing power of the natural gas pipeline network during time period t.

[0045] 7) Other operational constraints: The amounts of power outage load, heat outage load, and wind and solar curtailment must all be effectively controlled within a reasonable range, i.e., the following must be met: (44); (45); (46); In the above formula: , , These are the maximum power outage load, heat outage load, and wind / solar curtailment ratio, respectively. Let t be the electrical load during time period t; The heat load for time period t; Let t be the wind power output during time period t; This represents the power loss of the load during time period t; This represents the heat loss load power during time period t; This represents the amount of wind and solar power curtailed during time period t.

[0046] Step 6: Using the MATLAB simulation environment with the CPLEX optimization solver, input the source-load time series data and equipment parameters of the park, and solve the full life cycle health sensing multi-energy optimization model of Step 5 to obtain the number of modules, capacity, and load allocation strategy. Specifically, this includes: S6.1: Standardize the source load data and equipment parameters of the park to provide an input basis for model solving and ensure the consistency between data and formula constraints. Specific operations are as follows: ①. Source-load time-series data processing: Input photovoltaic power output, wind power output, electrical load, and thermal load for 8760 hours throughout the year (data is time-series normalized, referencing typical daily power curves for each season), time step. The corresponding time period in the formula is t=1,2,...,8760.

[0047] ②. Equipment parameter input: Set key parameters according to patent table 1-3, including fuel cell, electrolyzer, economic parameters, constraint parameters, etc.

[0048] ③. Data format conversion: Organize time series data and parameters into MATLAB matrices to adapt to the linear programming input requirements of the CPLEX solver.

[0049] S6.2: Build the optimization model from step 5 in MATLAB, transforming the objective function and constraints into mathematical expressions to achieve the coupling of "cost-health-multi-energy flow". Details are as follows: ①. Definition of objective function: based on equation (19): In MATLAB, the annualized total lifecycle cost is defined as the optimization objective, and each cost item is calculated according to the corresponding formulas, namely (20)-(25): ②. Constraint embedding: Using MATLAB's optimization toolbox, the constraints in step 5 are transformed into an inequality / equality constraint matrix, including power balance constraints, health state constraints, hydrogen energy storage constraints, decision variable constraints, etc. S6.3: Configure the CPLEX solver parameters, start the solution process, and perform the optimal solution search for the decision variables (number of modules, capacity, load allocation). Details are as follows: Solver parameter settings: In MATLAB, call the cplexlp function and configure the key parameters: Solution objective: Minimize ; Solution accuracy: Feasibility tolerance 1e 6. Optimize tolerance 1e 8; Algorithm selection: Linear programming (LP) algorithm is adopted because the cost terms and constraints in the model are all linear relationships (the nonlinearity of health degradation is transformed into linear constraints through the discretized SOH evolution model). Iteration limit: Set the maximum number of iterations to 10,000 to avoid solving timeout.

[0050] Model solution startup: Run the MATLAB script. CPLEX traverses the feasible region of decision variables using the branch and bound method, verifies constraint satisfaction time by time using time series data, and finally outputs the optimal solution, including: configuration parameters, number of fuel cell modules, rated capacity of a single module, and capacity of electrolyzer / hydrogen storage tank. Operating parameters: load allocation strategy for each time period, power purchase, hydrogen mixing power, and SOH evolution trajectory of each module.

[0051] S6.4: Verify the constraint satisfaction and rationality of the optimal solution, output standardized results, and provide a basis for subsequent scheme comparison, as follows: a. Constraint satisfaction verification: Verify the energy balance of electrical energy, thermal energy, and gas energy in each time period; Verify the health status of the module; Verify security constraints: b. Output of results: Configuration scheme: number of output modules, capacity of each device; Economic and performance indicators: Output annualized total life cycle cost, carbon emission reduction rate, and renewable energy integration rate; Operation trajectory: Outputs the load distribution time-series curves of each module, the SOH evolution curve, and the hydrogen storage SOC change curve, intuitively displaying the optimization effect.

[0052] This invention provides a method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization. The technical effects are as follows: 1) Step 1 of this invention establishes a fuel cell health status perception model, which can accurately describe the health status decay process based on the actual output and cumulative load of the equipment. It transforms the originally implicit and difficult-to-quantify life loss into a calculable and constrainable health indicator, providing a real and reliable basis for subsequent optimization and avoiding configuration deviations and life risks caused by ignoring aging characteristics in traditional methods.

[0053] 2) Step 2 of this invention proposes a multi-module health collaborative operation strategy, which dynamically allocates output according to the real-time health status of each module, controls the health differences between modules, realizes the matching operation of load and health status, effectively slows down the overall degradation speed of equipment, extends the balanced life of the system, improves the reliability and safety of operation, and solves the problems of uneven life and premature failure that are easy to occur in multi-module operation.

[0054] 3) Step 3 of this invention constructs a complete energy conversion and energy balance model for the hydrogen energy storage unit, clearly describing the electrical... hydrogen The study of energy flow, efficiency loss, and coupling relationships between heat sources, along with the standardization of energy measurement, provides an accurate physical basis for system-level multi-energy flow optimization. This enables subsequent configuration and scheduling to better align with actual engineering characteristics, thereby improving the model's rationality and applicability.

[0055] 4) Step 4 of this invention establishes a full life cycle cost optimization model, which integrates investment costs, operating costs, carbon emission costs, penalty costs, and health degradation costs into the economic evaluation. It adopts the equal annual value method to achieve reasonable conversion of long-term costs, breaks through the limitation of traditional configuration that only focuses on initial investment, and truly achieves economic optimization throughout the entire life cycle.

[0056] 5) Step 5 of this invention constructs a comprehensive optimization model that integrates health perception, multi-energy flow balance, and full life cycle cost. It organically combines equipment life constraints, energy supply and demand constraints, safe operation constraints, and economic goals. Under the premise of satisfying the balance of electricity, heat, and gas, it simultaneously achieves the lowest cost, balanced life, low carbon and high efficiency, forming a complete and innovative optimization system.

[0057] 6) Step 6 of this invention uses MATLAB and CPLEX solvers to efficiently solve the proposed model, which can handle large-scale time series data and complex optimization problems with multiple constraints. It can stably output the optimal equipment capacity, number of modules and load allocation strategy, with high solution accuracy and reliable results. It can provide a directly usable configuration scheme for engineering practice and has strong practicality and scalability.

[0058] 7) This invention addresses the problems in existing integrated energy system hydrogen energy storage configurations, such as neglecting differences in fuel cell health status, difficulty in adapting operating strategies to dynamic degradation, and insufficient connection between total life cycle cost and short-term operational optimization. It proposes a hydrogen energy storage configuration method that integrates health status perception and total life cycle optimization. The invention verifies the significant advantages of the proposed method in reducing annualized total life cycle cost, improving energy utilization efficiency, promoting renewable energy consumption, delaying the overall aging of fuel cells, and enhancing low-carbon performance. Attached Figure Description

[0059] The present invention will be further described below with reference to the accompanying drawings and examples; Figure 1 This is a graph showing the relationship between the state of health (SOH) of each module in a fuel cell and the cumulative load.

[0060] Figure 2 This is a graph showing the relationship between the remaining lifespan of each module in a fuel cell and future load requirements.

[0061] Figure 3 This is a diagram of a hydrogen energy storage architecture for a comprehensive energy system.

[0062] Figure 4(a) shows the power curves for a typical day (spring). Figure 4(b) shows the power curves for various types of power on a typical day (summer). Figure 4(c) shows the power curves for a typical day (autumn). Figure 4(d) shows the power curves for a typical day (winter).

[0063] Figure 5(a) is a schematic diagram of the health status perception operation strategy (initial state); Figure 5(b) is a comparison of group scheduling under the health perception strategy; Figure 5(c) is a comparison chart of cumulative aging losses of modules.

[0064] Figure 6(a) is a schematic diagram of the impact of the traditional equal distribution strategy on the dynamic evolution of the module health status; Figure 6(b) is a schematic diagram illustrating the impact of the health status perception strategy on the dynamic evolution of the module's health status.

[0065] Figure 7 Figure showing the system operating power and SOH response characteristics under optimized configuration throughout the entire lifecycle.

[0066] Figure 8(a) is a value decomposition diagram of the whole life cycle optimization configuration; Figure 8(b) is a chart showing the return on investment analysis of the whole life cycle optimization configuration.

[0067] Figure 9 This is a topology diagram of a multi-energy coupled system for hydrogen energy storage. Detailed Implementation

[0068] A comprehensive energy system hydrogen storage optimization configuration method based on health status perception and full life cycle optimization First, focusing on fuel cells, the core device of hydrogen energy storage systems, and based on their physical degradation mechanism, we construct a health state index decay model that considers the impact of cumulative load. We also design a module life consumption rate formula, an overall health state weighted average index, and future load carrying capacity constraints to accurately quantify the aging degree of multiple modules during long-term operation, laying the foundation for subsequent optimization. Secondly, to address the shortcomings of traditional operation strategies that ignore the differences in module health, a multi-module collaborative operation strategy based on real-time health status is designed, which includes module operation range constraints, SOH evolution model, health difference constraints between modules and quantitative indicators of remaining load capacity. Through dynamic power allocation and lifetime balancing mechanisms, the overall aging of the system is proactively delayed. Then, the research was extended to the integrated energy system of electricity, heat and gas, a hydrogen energy storage unit consisting of an electrolyzer, a hydrogen storage tank and a fuel cell was built, an electrochemical and thermochemical energy conversion model was established, a system-level multi-energy flow collaborative management model was constructed, and multi-energy efficient coupling was achieved.

[0069] Finally, by integrating the health status perception model, multi-module collaborative operation strategy, and multi-energy flow collaborative management model, with the core objective of minimizing the annualized total life cycle cost, and incorporating life cycle cost items such as equipment investment, operation and maintenance, and health degradation, a comprehensive energy system hydrogen storage optimization configuration model is established, covering constraints such as energy balance, system equipment operation, system pipeline network, and hydrogen blending ratio. The model achieves the adaptation of long-term cost and annual time-series simulation through the equal-annual value method.

[0070] Example: (a) Example parameters: Based on step 4, a full life cycle optimization configuration model is established, and the parameters of the fuel cell are shown in Table 1.

[0071]

[0072] Based on step 5, a multi-functional optimization model for health perception throughout the entire life cycle is established, and the parameters are shown in Tables 2 and 3.

[0073]

[0074]

[0075] During system operation, the maximum allowable power load and heat load loss ratio is 5%, and the penalty for load loss is set at 10 times the real-time electricity price. Simultaneously, the maximum allowable wind and solar power curtailment ratio is 10%, with a penalty coefficient of 0.2 yuan / (kW·h) for curtailment. Furthermore, the maximum hydrogen mixing ratio in the system is 20%. The carbon emission cost is 250 yuan / t, the carbon emission factor for electricity is 40 kg / (MW·h), while the carbon emission factor for natural gas is calculated based on methane data.

[0076] According to step 6, based on actual monitoring data from an industrial park, key parameters such as wind power output, photovoltaic power generation, electrical load, and thermal load were selected and time-series normalized as the benchmark dataset. A year-round cycle was then used to study the optimal configuration of the hydrogen energy storage system. The park's load exhibits significant seasonality, with high electrical load in summer and high thermal load in winter, and the supply of clean energy also fluctuates seasonally. Typical daily scenarios for each season (specific typical daily power distributions are shown in Figures 4(a) to 4(d)) were used to verify the applicability of the optimization model and the adjustment capability of the hydrogen energy storage system. These typical daily loads and the volatility of renewable energy provide necessary conditions for evaluating the adaptability of fuel cell health management strategies and life-cycle configuration strategies under different seasonal operating conditions.

[0077] (II) Optimization Result Configuration Analysis: This invention sets up three comparative schemes to analyze the advantages of the proposed collaborative configuration and operation strategy for hydrogen energy storage systems based on health status perception and full life cycle optimization. The specific schemes are as follows: Case 1: The number of fuel cell modules is fixed, and the operation of the modules is managed only by time, without considering health status and lifespan degradation.

[0078] Case 2: The number of fuel cell modules is fixed. A health status perception model is introduced to dynamically allocate the load based on the health status of each module.

[0079] Case 3: The number of fuel cell modules can be adjusted, and a health status perception and multi-module collaborative optimization strategy is combined to optimize the total cost over the entire life cycle.

[0080]

[0081] As shown in Table 4, the fuel cell health status perception and full life cycle optimization strategy proposed in this invention can effectively reduce the overall energy cost of the industrial park. Specifically, compared with the baseline scheme of Case 1, Case 2, which introduces health status perception, has reduced the annualized full life cycle cost by RMB 3988.83 through dynamic load allocation. This is mainly due to the improvement in operating efficiency and the saving of energy purchase costs. Furthermore, the Case 3 scheme, which adopts an adjustable number of modules and targets the total life cycle cost, further reduces the total cost to RMB 171884.641, saving an additional RMB 2118.86 compared to Case 2, making it the optimal configuration among the three schemes. The superior performance of Case 3 proves that explicitly incorporating the health degradation and life loss of fuel cells into the LCC optimization objective can guide the system to make globally optimal configuration decisions, thereby achieving the best economic balance between initial investment, operating efficiency, and long-term maintenance.

[0082]

[0083] Table 5 shows the configuration results, reflecting the trade-offs between the optimization strategy and the system's decarbonization and economic objectives. The electrolyzer capacity remains constant across all schemes, reflecting that its configuration scale is limited by the park's power structure or maximum hydrogen production demand. The significant reduction in fuel cell and hydrogen storage tank capacities reveals the core driving force of the optimization model. From Case 1 to Case 2, the fuel cell configuration capacity decreased from 52.88kW to 39.95kW, indicating that the system successfully replaced the traditional static capacity reserve requirement by introducing module performance evaluation and dynamic load allocation mechanisms, thus ensuring operational reliability and efficiency with a smaller installed capacity. In the final scheme, Case 3, the configuration capacity was further compressed to 30.00kW for the fuel cell and 115.70kW for the hydrogen storage tank. The fundamental reason is that the optimization objective shifted from simply meeting reliability requirements to minimizing the overall life-cycle cost. High capital expenditure equipment is strictly constrained under this objective function, forcing the system to seek the minimum investment to meet energy consumption demands. It is worth noting that the significant capacity reduction in Case 3 does not imply a decrease in reliability, but rather a redefinition of system redundancy. Traditional methods often over-configure to cope with uncertainty by ignoring lifespan decay, resulting in idle and wasted assets. This strategy, however, finds the optimal economic balance between capacity and lifespan by quantifying lifespan cost, accurately eliminates ineffective redundant capacity, and utilizes intelligent scheduling strategies based on SOH awareness to tap the potential of equipment, thereby ensuring operational safety and long lifespan under small capacity configuration.

[0084]

[0085] The comprehensive comparison results in Table 6 clearly verify the superiority of the proposed multi-objective optimization configuration strategy in the planning of the park's integrated energy system. Taking Case 1 as a reference, under the core objective of minimizing the annualized total life-cycle cost, the low-carbon operation requirements are achieved synergistically. Scheme Case 3 achieves a comprehensive improvement in both system economy and environmental benefits. From an economic perspective, although higher-cost low-carbon technologies are introduced, through precise coupling of capacity configuration and operation strategies, the annualized total life-cycle cost of Case 3 is still controlled at RMB 171,884.64, achieving a total cost saving rate of 3.43%, indicating that the optimization strategy effectively overcomes the inherent conflict between cost and emission reduction targets. From an environmental perspective, the annual carbon emissions of Case 3 are effectively reduced to 41.15 tons, with a carbon reduction rate as high as 47.24%. This significant emission reduction effect is the result of the system configuration tilting towards lower-carbon and more flexible energy conversion and storage units under the guidance of multi-objective optimization. This optimized configuration significantly improves the overall energy utilization efficiency of the system and achieves a renewable energy consumption level of 95.00%.

[0086] (III) Modular fuel cell operating characteristic analysis based on health status perception: To quantitatively evaluate the actual performance of the health status-aware operation strategy proposed in this invention, this section compares the traditional equal-distribution strategy used in Case 1 with the health status-aware strategy through simulation. The core purpose of the simulation is to verify the superiority of the new strategy in balancing aging losses among modules, thereby extending the overall system lifespan. The initial health status of each module, the power scheduling situation during typical periods, and the cumulative aging losses after a complete operating cycle are compared under the two strategies, as shown in Figures 5(a) to 5(c).

[0087] The optimization effect of the health status-aware operation strategy on the lifespan balancing of modular fuel cells is shown in Figures 5(a) to 5(c). Figures 5(a) to 5(c), through comparative analysis of the system, reveal the inherent law between the operation strategy and equipment lifespan degradation, providing an important decision-making basis for the capacity planning of hydrogen energy storage systems. Figure 5(a) presents the health status distribution of each fuel cell module at the initial moment, showing significant performance differences between modules. If this initial state characteristic is ignored in system planning and a uniform load distribution method based on fixed-sequence rotation is adopted, it will lead to the non-equilibrium aging phenomenon shown in Figure 5(c) after long-term operation: modules with weaker initial health status exhibit a faster performance degradation rate. This trend of locally accelerated aging will force the system to replace equipment prematurely to maintain operational reliability, thus significantly increasing the capacity replacement cost throughout the entire life cycle. In contrast, the health status-aware operation strategy proposed in this invention, as shown in Figure 5(b), achieves effective coordination of the aging process between modules by establishing a power distribution mechanism adapted to the health status, as shown in Figure 5(c). This strategy, through operational-level optimization and control, significantly slows down the overall system performance degradation process, thereby extending the effective service life of the configured capacity. This means that, under the same service life requirements, the design redundancy or equipment replacement frequency required by the system can be effectively controlled.

[0088] The above analysis verifies the effectiveness of the health status awareness strategy in balancing instantaneous aging losses from a static perspective. To further reveal the impact of this strategy on the long-term reliability of the system, it is necessary to track the dynamic evolution of the module's health status. The following will show the module's SOH evolution trajectory over a period of one year under different strategies, revealing the long-term value of the strategy in delaying the overall performance degradation of the system from a dynamic dimension, as shown in Figures 6(a) and 6(b).

[0089] The impact of different operating strategies on the dynamic evolution of module health status is shown in Figures 6(a) and 6(b). Under the traditional equal-weighting strategy shown in Figure 6(a), the SOH trajectory of the modules exhibits significant divergence due to the inherent initial inconsistencies and differences in decay sensitivity among the modules within the system. Specifically, modules with lower initial SOH, continuously bearing the same average load as other modules, experience a more pronounced accelerated decay effect, resulting in a steeper decline in their SOH curves. Ultimately, they are the first to fall below the failure threshold after approximately 4.5 months of operation, leading to premature retirement or performance limitations of the entire system. In contrast, the health status-aware strategy shown in Figure 6(b) demonstrates superior lifetime balancing capabilities. This strategy implements differentiated power allocation and load management by real-time monitoring and evaluation of the SOH of each module, combined with advanced predictive models. It dynamically allocates relatively larger loads to modules with higher health status while limiting the load or implementing hibernation protection for modules with lower health status to slow their decay, thereby achieving precise control over the module SOH decay rate. The results show that by the end of the 12-month operation cycle, the SOH of all modules remained stable above 0.7, and the maximum SOH difference between modules was controlled within 0.05. This achieved the equalization of module lifespan, that is, the SOH difference between modules was significantly suppressed. This strongly demonstrates that the strategy can eliminate the negative impact of system inconsistency on lifespan and significantly improve the distribution characteristics of system operating lifespan.

[0090] (iv) Analysis of operational strategies under full lifecycle optimized configuration: After completing the operational characteristic analysis of the modular fuel cell with health status awareness, the effectiveness of the optimized configuration strategy is further verified from a full life cycle perspective. Unlike traditional configuration methods that only consider single-day or short-term operation, full life cycle optimization not only focuses on the comprehensive costs of investment, operation, degradation, and maintenance throughout the entire life cycle of the equipment, but also prompts the system to proactively avoid future health degradation risks during the configuration phase, thereby achieving higher economic efficiency and reliability. Based on this, the power distribution and SOH dynamic response characteristics of the system on a typical day under full life cycle optimized configuration are first presented to illustrate how the optimized system structure achieves smoother operation and a healthier degradation path. Specifically, as shown below... Figure 7 As shown: The system operating power and SOH response characteristics under full lifecycle optimized configuration are shown in the figure. Figure 7 As shown, Figure 7This demonstration showcases the response characteristics of electrolyzer power consumption, fuel cell power, and hydrogen storage system state of charge (SOC) under a fully optimized lifecycle configuration over a 24-hour period. From 0 to 8 hours, the electrolyzer power fluctuates smoothly, while the fuel cell power remains at a low level, and the SOC of the hydrogen storage system remains stable in the 0.4-0.6 range. The equipment operating intensity exhibits mismatched regulation, avoiding the sustained high-load operation seen in traditional initial investment-driven strategies. From 8 to 20 hours, the fuel cell power gradually climbs to its peak, while the electrolyzer power does not simultaneously experience high-load output. The hydrogen storage SOC adjusts orderly to 0.2 before rebounding, without extreme overcharging or over-discharging states. The depth of charge / discharge is constrained, reducing the risk of accelerated equipment wear. The dynamic adaptation of multiple equipment power curves to the hydrogen storage SOC demonstrates that lifecycle optimization quantifies equipment lifespan losses upfront during the planning stage and achieves real-time control of losses through operational strategies, breaking down the barriers between planning and operation in traditional models. This configuration achieves orderly control of losses by dynamically balancing equipment workload, aligning with the optimization objective of optimal total lifecycle cost, rather than the single-dimensional optimization of initial investment.

[0091] Figure 7 From the perspective of the dynamic characteristics of the operation process, it is explained that the full life cycle optimized configuration achieves balanced management of equipment losses through the coordinated regulation of equipment power and hydrogen storage SOC. This is the technical implementation of this configuration at the operation end. The optimal life cycle cost of this configuration needs to be verified through quantitative investment and return relationship. As shown in Figure 8(a) and Figure 8(b), the strategic investment and return decomposition provides a direct quantitative representation of its economic value.

[0092] The value decomposition and return on investment analysis of the full life cycle optimized configuration are shown in Figures 8(a) and 8(b). Figures 8(a) and 8(b) quantify the value transformation relationship of the full life cycle optimized configuration through strategic investment and return decomposition and return on investment analysis: The decomposition results in Figure 8(a) show that the return of this configuration is centered on energy purchase savings, supplemented by investment savings and penalty savings, reflecting that its value does not rely on short-term cost reduction, but rather on optimizing the operating cost item with the highest proportion throughout the entire life cycle. Figure 8(b) further clarifies that the strategic investment of 2,800 yuan corresponds to a total return of 10,107 yuan, with a return on investment of 3.61:1. This result confirms that this configuration achieves orderly control of losses by dynamically balancing the working intensity of equipment and combining load distribution based on health status perception. This avoids the over-configuration of traditional solutions and solves the limitation of focusing only on perception while ignoring long-term costs.

Claims

1. A method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization, characterized in that... Includes the following steps: Step 1: Establish a fuel cell health status perception model using an exponential decay method; Step 2: Based on the fuel cell health status perception model built in Step 1, formulate a multi-module collaborative operation strategy to address the dynamic differences in module health. Step 3: Construct an energy conversion model for the hydrogen energy storage unit using a unified energy index; Step 4: Based on the adaptability between long-cycle costs and short-term operation, establish a full lifecycle optimization configuration model; Step 5: Based on the hydrogen energy storage unit energy conversion model in Step 3 and the full life cycle optimization configuration model in Step 4, integrate the multi-energy coupling requirements of electricity, heat and gas to construct a full life cycle health perception multi-energy optimization model; Step 6: Input the source-load time series data and equipment parameters of the park, solve the full life cycle health perception multi-energy optimization model in Step 5, and obtain the number of modules, capacity and load allocation strategy.

2. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 1, characterized in that: It also includes step 7: setting up multiple comparison schemes, and comparing and analyzing the annualized cost, carbon emission reduction rate, renewable energy consumption rate and module SOH evolution trajectory corresponding to the solution results of step 6 under typical day scenarios in the four seasons.

3. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 1, characterized in that: In step 1, based on the physical degradation law of fuel cells and combined with the impact of cumulative load on lifespan, a fuel cell health status perception model is established using an exponential decay method. The specific construction method is as follows: （1）； In formula (1): This represents the percentage of the fuel cell module i in terms of its health status; L represents the cumulative output power. This represents the maximum load during the planning period. Specific attenuation coefficient for the module; Indicates the total number of fuel cell modules; This is the exponential decay control parameter. A value greater than 1 indicates a nonlinear amplification effect of load intensity on health decline; The degradation process is described using a module lifetime consumption rate formula, as shown below: （2）； In formula (2): This indicates that the i-th fuel cell module is under the cumulative equivalent load The health decay rate under certain conditions is used to describe the trend of the module's state of health (SOH) as the operating load accumulates; Represents the cumulative load L The power is used to describe the nonlinear change in degradation rate with increasing operating load; In a multi-module system, a weighted average indicator is introduced to evaluate the overall health status, as shown below: （3）； In formula (3): This represents the weighted average of the overall health status of a multi-module fuel cell system, with a value range of [0,1]. The module weight is determined based on the importance of the module or the expected load allocation. This indicator can reflect the health level of the entire fuel cell stack. Furthermore, to ensure that the configuration scheme can meet future load demands in the long term, the constraint of the current remaining lifespan on the carrying capacity of future loads should be considered: （4）； In equation (4): This represents the equivalent cumulative load that the module will need to bear in the future during the planning period; Indicates the current remaining lifespan percentage; This indicates that the current accumulated load of the i-th fuel cell module is... The proportion of people in a healthy state at that time; This indicates the current accumulated operating load of the module.

4. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 3, characterized in that: In step 2, to prevent modules with low SOH from continuing to bear high loads and thus accelerating degradation, the module... i During the period t The allowed operating range constraints are as follows: （5）； In equation (5): Representation module i During the period t The actual output power is the optimization decision variable, representing how much load the system ultimately decides to let the module bear at that moment; Representation module i The rated maximum output power that can be provided when SOH=1; Representation module i During the period t health status, ; Secondly, in order to dynamically track the degradation process of each module in the optimization model, an evolution model of SOH with operating load is constructed. Based on the high load-high degradation characteristics of fuel cells, the degradation expression driven by load intensity is as follows: （6）； In formula (6): Representation module i In the next period t+ 1. Health status; Representation module i During the period t ; output power; Representation module i Rated power; Indicates the coefficient of health decline; Indicates the optimization time step; The amplification effect of high load on the sensitivity to decline is described; Equation (6) is a discretized approximation of Equation (1) and is used for annual time-series operation simulation.

5. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 4, characterized in that: To avoid an uneven lifespan distribution caused by some modules operating under heavy loads for extended periods while others remain idle, a health deviation constraint between modules is further introduced. This constraint guides optimization to tilt the load towards modules with higher health levels, while preventing the health deviation from widening further. The health deviation constraint between modules is expressed as follows: （7）； In equation (7): Representation module i With modules j During the period t Health deviations; This represents the health status ratio of the i-th fuel cell module in time period t, with a value range of [0,1]. It is a real-time quantitative indicator of the health status of a single module. This represents the percentage of the j-th fuel cell module in good health during time period t; This represents the maximum permissible threshold for health differences. Represents the discretized scheduling time; Indicates the total number of time periods within the planning period; Within the framework of lifecycle optimization, relying on the health degradation cost in the objective function. To flexibly guide the lifespan balance among modules; the constraints on the health differences among modules serve as auxiliary boundary conditions. Building upon this, a future availability forecast based on health status is introduced to achieve a quantitative assessment of module lifespan sustainability, as detailed below: （8）； In equation (8): Representation module i During the period t The remaining load capacity is used to quantify its sustainable operating capacity over a period of time.

6. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 5, characterized in that: In step 3, combining the bidirectional electro-hydrogen-thermal conversion characteristics and based on electrochemical and thermochemical principles, a unified energy index is used to construct an energy conversion model for the hydrogen energy storage unit; the specific construction method is as follows: (1) Electrochemical characteristics: In the hydrogen energy storage unit, the electrolyzer and the fuel cell are used to produce hydrogen by consuming electricity and generate electricity by burning hydrogen, respectively. The hydrogen-to-electricity conversion models for the two are shown in Equation (9) and Equation (10), respectively: （9）； （10）； In the above formula: , They are respectively t Power consumption and hydrogen production capacity of the time-phase electrolyzer; , They are respectively t The power consumption and hydrogen production capacity of the fuel cell during different time periods; and These are the energy conversion efficiencies of hydrogen production using electricity consumed by an electrolyzer and hydrogen power generation using a fuel cell, respectively. A unified energy index is used to characterize the dynamic conversion process of electrical energy, thermal energy and hydrogen energy in the hydrogen energy storage unit; based on the principle of energy equivalent conversion, the hydrogen storage capacity is converted into standard energy units through its calorific value parameter. Specifically, under standard conditions of 0℃ and 1 atmosphere, the lower heating value of hydrogen per unit volume is taken as 2.95 kW·h / m³, thereby realizing the equivalent conversion between hydrogen physical quantities and energy values. (2) Thermochemical properties: The thermodynamic properties of electrolyzers and fuel cells are characterized by the following equations: （11）； （12）； In the formula: The operating temperature of the electrolyzer during time period t is the electrolyte temperature. The heat capacity of the electrolytic cell; The operating temperature of the fuel cell during time period t; The heat capacity of the fuel cell; , These represent the heat absorption power of the cooling circulating water in the electrolyzer and fuel cell during time period t, respectively. Let t be the power consumption of the electrolytic cell during time period t; The hydrogen production power of the electrolyzer during time period t; The hydrogen consumption power of the electrolyzer during time period t; Let t be the power consumption of the electrolytic cell during time period t.

7. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 6, characterized in that: In step 4, the fuel cell health status perception model from step 1 is integrated with the multi-module collaborative operation strategy from step 2. Based on the adaptability between long-term costs and short-term operation, a health degradation cost discounting mechanism is incorporated to establish a full life-cycle optimization configuration model. The specific construction method is as follows: Based on the equal annual cost (EAC) method, a cost quantification model is first constructed for the entire life cycle, and then converted into an annualized life cycle cost for subsequent modeling and solving; among them, the long-term replacement cost caused by the health degradation of fuel cells will be allocated to each operating year to form the corresponding annualized health degradation cost. The objective function for the total lifecycle cost of the system is as follows: （13）； In equation (13): This represents the total lifecycle cost of the fuel cell subsystem. N This represents the total number of fuel cell modules; For module i One-time investment cost; , , and Modules i Future load during the planning period The calculated operating and maintenance costs, calculated carbon emission costs, calculated operating penalty costs, and calculated energy purchase costs are as follows: Representation module i Future period t The equivalent load to be borne; To incorporate health status into configuration optimization, the constraints on system capacity from the future equivalent load requirements of modules are as follows: （14）； In equation (14): For module i Rated output capacity; To configure the initial module health status ratio; function Used to quantify the current health status's limitation on future tolerable loads; The choice of module capacity and quantity affects future load capacity and cost consumption. The capacity constraints are introduced as follows: （15）； In equation (15): This represents the maximum load demand during the planning period; To comprehensively consider lifespan depletion and economic efficiency, a health degradation cost is defined. This cost converts the module's lifespan loss into quantifiable economic depreciation and is explicitly incorporated into the objective function; its calculation formula is shown below: （16）； In equation (16): It is expressed as the total cost of health degradation over the entire life cycle, which is the total replacement cost corresponding to the life loss of the equipment throughout its entire service life. This represents the cost of replacing the i-th fuel cell module once. and These are the module's health status at the beginning and end of the planning period, respectively; The health threshold at the end of life; The discount factor is given by equation (17). （17）； In equation (17): This represents the annual discount rate, which is the time value of money rate. Indicates the period number in which the future cost will occur; for t The discount factor for a given period is used to discount future replacement costs to their present value. Based on formula (16), i.e. According to the service life of the equipment m With interest rates r Convert it into an annualized cost of health degradation. This is incorporated into the annual optimization goals, thus linking the lifecycle loss with the annual operating strategy. The final configuration optimization problem is formalized as follows: （18）； In equation (18): The set of decision variables for module capacity; N To optimize the number of modules in the object; This represents the minimum capacity of the i-th module; This represents the maximum capacity of the i-th module; This represents the minimum number of fuel cell modules; This represents the maximum number of fuel cell modules; in this model Characterize the total cost of the fuel cell subsystem throughout its entire lifecycle.

8. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 7, characterized in that: Step 5 includes setting constraints: Based on total life cycle cost Total cost of health degradation To balance the optimization objectives of the entire lifecycle with the solution requirements of annual time-series simulations, the cost is transformed into an annualized comprehensive lifecycle cost using the equal-annual value method. This is then used as the objective to achieve coordinated optimization of the hydrogen energy storage system's configuration and operation. Specifically, each cost is transformed into an annual cost using the following two annualization methods: ①: For one-time costs such as initial investment and total life-cycle health degradation costs: the equivalent annual value method is used, i.e., through the capital recovery factor. Converted to annual equivalent costs, of which, Indicates the service life of the equipment; ②: For operating costs such as energy purchase costs, carbon emission costs, penalty costs, and operation and maintenance costs that occur in different time periods: directly take the sum of their annual values, which is the reflection of the total life cycle time period costs in a single year.

9. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 8, characterized in that: The full life-cycle health perception multi-functional optimization model includes the following objective function: Of the two methods: Equations (20) and (25) are the first method; Equations (21), (22), (23), and (24) are the second method; both methods correspond to the full lifecycle cost in step 4, ensuring the compatibility between long-term optimization objectives and short-term time-series simulations, as shown in Equation (19): （19）； （20）； （21）； （22）； （23）； （24）； （25）； In the above formula, This represents the annualized total life-cycle cost of hydrogen energy storage configuration in an integrated energy system of electricity, heat and gas. The smaller the value, the better the long-term economic performance of the system. , , , , and These represent annualized investment cost, annualized energy purchase cost, annualized carbon emission conversion cost, annualized operating penalty cost, annualized operation and maintenance cost, and annualized health degradation cost, respectively. Among them: annualized investment cost Compared with the unit capacity investment cost of electrolyzers, fuel cells, and hydrogen storage tanks , , and configuration capacity , , Related; Annualized energy purchase cost Depend on t Time-of-use electricity pricing Gas prices With power purchase Gas purchase volume rate Combined with unit duration and number of time periods throughout the year calculate; Annualized carbon emission cost of , These are the carbon emission costs per unit of electricity and per unit volume of natural gas, respectively, obtained by multiplying the carbon emission price by the carbon emission factor. Annualized operating penalty cost Involving t Periodic power outage load Heat loss load Wind and solar power curtailment And the unit price for power outages and heat loss load penalties , and wind and solar curtailment penalty coefficient ; Annualized operating and maintenance costs With electrolyzers, fuel cells and hydrogen storage tanks t Time-based unit power operation and maintenance cost , , and operating power , , related; In this model Minimizing is equivalent to total lifecycle cost The minimization of the two, and their quantification relationship is shown in equation (26), that is... yes Equivalent decomposition at the annual level; （26）； In the above formula, The total lifecycle cost of a fuel cell subsystem is the sum of fixed and variable costs throughout the entire lifecycle of the equipment, from commissioning to decommissioning. This represents the total cost of health degradation over the entire lifecycle of the fuel cell subsystem.

10. The method for optimizing the configuration of hydrogen energy storage in a comprehensive energy system based on health status perception and full life cycle optimization as described in claim 9, characterized in that: The objective function includes constraints: 1) Power balance constraints: （27）； In equation (27): , They are respectively t Solar and wind power output during different time periods; for t Electrical load during a given time period; for t The power generation capacity of the gas turbine during a given period; This represents the amount of wind and solar power curtailed during time period t; This represents the amount of electricity purchased from the external power grid during time period t; This represents the power loss of the load during time period t; This represents the power consumption of the electrolytic cell during time period t; This represents the power generation of the fuel cell during time period t; 2) Thermal energy balance constraint: In the IES's thermal energy supply architecture, the local heat load is shared by the gas turbine unit and the hydrogen energy system, which satisfies: （28）； In equation (28): , They are respectively t Heat production capacity of gas turbines and gas boilers during specific time periods; for t Heat load during a given period; This represents the heat output power of the fuel cell during time period t. This represents the heat generation power of the electrolytic cell during time period t; This represents the heat loss load power during time period t; 3) Gas energy balance constraint: As shown in equation (29), the gas supply for Class 2 gas equipment is provided by purchased natural gas and hydrogen mixed into the pipeline network; （29）； In equation (29): This represents the rate at which gas is purchased from the external gas network during time period t. The calorific value of natural gas; , They are respectively t Gas consumption of gas turbines and gas boilers during specific time periods; The hydrogen mixing power of the natural gas pipeline network during time period t 4) Operating constraints of gas equipment: The electro-thermal energy conversion relationship between gas turbines and gas boilers, as well as the upper and lower limits of their output, are shown in equations (30) to (349): （30）； （31）； （32）； （33）； （34）； In the above formula: , These are the power generation efficiency and waste heat utilization efficiency of the gas turbine, respectively. The heat production efficiency of a gas-fired boiler; , These are the maximum power generation capacity of the gas turbine and the maximum heat production capacity of the gas boiler, respectively. This represents the power generation capacity of the gas turbine during time period t; This represents the gas consumption of the gas turbine during time period t; This represents the gas consumption of the gas-fired boiler during time period t. 5) Operational constraints of hydrogen energy storage units: The upper and lower limits of power for electrolyzers and fuel cells are shown in Equations (35) and (36), respectively, and the upper and lower limits of SOC for hydrogen storage and SOC for thermal storage are shown in Equations (37) and (38), respectively. （35）； （36）； （37）； （38）； In the above formula: This represents the power consumption of the electrolytic cell during time period t; Indicates the rated capacity of the electrolytic cell; This represents the power generation of the fuel cell during time period t; Indicates the rated capacity of the fuel cell; This represents the state of charge (SOC) of hydrogen storage during time period t. Let t be the state of charge (SOC) of hydrogen storage. Let SOC be the thermal storage state at time t; , and , These are the minimum and maximum values ​​of the hydrogen storage SOC and thermal storage SOC, respectively. During the configuration process, the SOC value remains unchanged at the beginning and end of the daily scheduling cycle, as shown in equations (39) and (40): （39）； （40）； In the formula: T The number of time periods in a daily scheduling cycle; This represents the initial value of the hydrogen storage SOC during the daily scheduling cycle; This represents the final state value of hydrogen storage SOC during the daily scheduling cycle; This is the initial value of the thermal storage SOC during the daily scheduling cycle; This represents the final state value of the thermal storage SOC during the daily scheduling cycle; 6) System piping constraints: Considering the physical constraints of the actual energy network, the electricity purchased from the external power grid and the gas purchased from the external gas grid both meet the following requirements: （41）； （42）； In the above formula: The power purchased from the external power grid; The gas purchase volume rate of the external gas network; , These are the maximum power purchase capacity of the external power grid and the maximum gas purchase volume rate of the external gas grid, respectively. In addition, considering the safety of the natural gas pipeline network and the combustion performance of terminal appliances, the volume of hydrogen mixed in natural gas must meet the following requirements: （43）； In equation (43): This refers to the calorific value of hydrogen. This represents the maximum hydrogen mixing ratio; This represents the hydrogen mixing power of the natural gas pipeline network during time period t; 7) Other operational constraints: The amounts of power outage load, heat outage load, and wind and solar curtailment must all be effectively controlled within a reasonable range, i.e., the following must be met: （44）； （45）； （46）； In the above formula: , , These are the maximum power outage load, heat outage load, and wind / solar curtailment ratio, respectively. Let t be the electrical load during time period t; The heat load for time period t; Let t be the wind power output during time period t; This represents the power loss of the load during time period t; This represents the heat loss load power during time period t; This represents the amount of wind and solar power curtailed during time period t.

Images