Comprehensive energy system double-layer optimization method and system considering equipment performance degradation and operation

Abstract

A kind of double-layer optimization method and system of comprehensive energy system considering equipment performance degradation and operation and maintenance, comprising: for each energy conversion device in comprehensive energy system, establish the equipment dynamic efficiency model coupling short-term variable working condition correction factor and long-term performance degradation correction factor regulated by maintenance level;Based on the equipment dynamic efficiency model of each energy conversion device, construct system energy conversion coupling model;Construct double-layer optimization model including planning layer and running layer;Planning layer minimizes system life cycle cost as target, outputs the configuration capacity and maintenance level of each equipment, running layer minimizes operating cost as target, based on the configuration capacity and system energy conversion coupling model regulated by maintenance level, construct constraint condition, output operating cost and feedback to planning layer;Solve model to obtain optimal scheme.The present application realizes the collaborative optimization of capacity configuration, operation scheduling and operation and maintenance strategy in the whole life cycle of electricity-hydrogen-heat comprehensive energy system.

Description

Technical Field

[0001] This invention relates to the field of planning and operation optimization technology for integrated hydrogen energy systems, and in particular to a two-layer optimization method and system for integrated energy systems that takes into account equipment performance degradation and operation and maintenance. Background Technology

[0002] Integrated energy systems (IES), characterized by the coupling of multiple energy sources including electricity, hydrogen, heat, and cooling, have become a key physical carrier for energy transition. However, existing IES optimization modeling and planning methods have significant limitations in practical applications, leading to optimization results that often deviate from the actual optimal solution, and even resulting in wasted investment or insufficient capacity throughout the entire life cycle. Existing technologies mainly suffer from the following defects: 1. The models are overly idealistic, ignoring the nonlinear and variable operating conditions of the equipment: Traditional energy hub models typically assume that the conversion efficiency of equipment (such as electrolyzers and fuel cells) is constant. Although some "Dynamic Energy Hub" (DEH) models consider the impact of load rate on efficiency through piecewise linearization, these models are mostly limited to short-term intraday scheduling and fail to reflect the performance changes throughout the entire life cycle of the equipment. 2. The correlation between long-term performance degradation and maintenance is ignored: The core equipment in hydrogen energy modules (proton exchange membrane fuel cells PEMFCs and electrolyzers PEMECs) are typical electrochemical devices, and their performance (such as voltage and efficiency) will irreversibly degrade with the accumulation of operating time. Most existing planning models assume that equipment performance remains constant over a 15-20 year lifespan, leading to an underestimation of later operating costs. 3. The disconnect between planning, operation, and maintenance strategies: Existing research often treats "capacity planning," "operation scheduling," and "equipment maintenance" as isolated issues. In reality, these three are strongly coupled, including: short-term system operation strategies (such as start-stop frequency and high-load operation time) accelerate equipment aging; the level of maintenance directly determines the aging rate of the equipment (better maintenance results in slower degradation, but higher maintenance costs); long-term performance degradation, in turn, changes the equipment's variable operating condition efficiency curve, thus affecting future optimal scheduling strategies.

[0003] Chinese patent application CN121660309A discloses a scheduling method for an integrated energy system. By constructing a lifespan decay model and cost quantification model for the entire life cycle of energy storage equipment, and integrating particle swarm optimization algorithm with a real-time dynamic adjustment mechanism, it incorporates the equipment lifespan loss cost into the core consideration of the total scheduling cost. However, it still has the following shortcomings: It lacks a two-level overall planning mechanism of "planning-scheduling". This invention remains at the level of operation scheduling under the given system parameters. Its scheme completely lacks a planning mechanism for optimizing the initial capacity of equipment on a long-term (such as the entire life cycle) scale, which leads to it being able to only "passively schedule based on existing equipment" and failing to fundamentally solve the pain point of the disconnect between the early capacity configuration and the later operational evolution of the system. When processing the generation of scheduling schemes, this invention directly uses particle swarm optimization algorithm to optimize the output allocation ratio of each energy supply device and the charging and discharging power of each energy storage device. When facing large-scale systems with deep coupling of multiple energy flows such as electricity, hydrogen, and heat, and with stringent operational constraints, simply relying on heuristic algorithms to directly solve complex scheduling problems across multiple time periods is prone to getting trapped in local optima and is computationally inefficient. The lifetime degradation model of this invention only considers factors such as the number of charge-discharge cycles and actual operating power as passive factors affecting lifetime loss. The system does not introduce the key active control variable of "maintenance level," resulting in it only being able to delay loss by "reducing equipment output." The lifetime loss and degradation calculation of this invention is mainly based on the state of charge and charge-discharge depth, which are typical characteristic models of traditional electrochemical energy storage. It does not specifically characterize the short-term variable nonlinear efficiency characteristics of hydrogen-containing energy systems (such as electrolyzers and fuel cells) under different load rates, making it difficult to accurately guide the scheduling of integrated energy systems containing a high proportion of hydrogen energy.

[0004] Currently, there is still a lack of a modeling method that can integrate long-term capacity planning, proactive operation and maintenance decisions, and short-term variable condition precise scheduling across different time scales into a two-layer architecture for collaborative optimization. Summary of the Invention

[0005] This invention aims to solve the above problems by providing an integrated operation and maintenance optimization method that takes into account the performance degradation and variable operating conditions of hydrogen energy cabin equipment. This invention constructs a dynamic equipment model that couples short-term variable operating conditions, long-term degradation, and maintenance level, and introduces maintenance level as a decision variable in a two-layer optimization architecture. This enables the coordinated optimization of capacity configuration, operation scheduling, and operation and maintenance strategies throughout the entire life cycle of the integrated electric-hydrogen-thermal energy system, in order to achieve the goal of the lowest overall system cost and the best reliability.

[0006] The present invention adopts the following technical solution.

[0007] In a first aspect, the present invention discloses a two-layer optimization method for a comprehensive energy system that considers equipment performance degradation and operation and maintenance, the method comprising the following steps: For each energy conversion device in an integrated energy system, a dynamic efficiency model is established that couples a short-term variable operating condition correction factor with a long-term performance degradation correction factor controlled by maintenance level. Based on the equipment dynamic efficiency model of each energy conversion device, a system energy conversion coupling model is constructed. A two-layer optimization model is constructed, consisting of a planning layer and an operation layer. The planning layer aims to minimize the total lifecycle cost of the system and outputs the configuration capacity and maintenance level of each device. The operation layer aims to minimize the operating cost and constructs constraints based on the system energy conversion coupling model regulated by the configuration capacity and maintenance level. It outputs the operating cost and the output of each device and feeds the operating cost back to the planning layer. Solving the two-layer optimization model yields the capacity configuration and operation scheduling scheme of the integrated energy system.

[0008] More preferably, The dynamic efficiency of the equipment is the product of the reference energy efficiency of the equipment under rated operating conditions and the short-term variable operating condition correction factor and the long-term performance degradation correction factor; the short-term variable operating condition correction factor is a function of the equipment load rate and is used to represent the dynamic fluctuation characteristics of the equipment efficiency as the load rate changes; the long-term performance degradation correction factor is a function of the equipment's operating years and maintenance level, and the higher the maintenance level, the larger the long-term performance degradation correction factor.

[0009] More preferably, The energy conversion devices include fuel cells, proton exchange membrane electrolyzers, and refrigeration units.

[0010] More preferably, Short-term variable operating condition correction factor for fuel cells ( ,in, This is a short-term variable operating condition correction factor for fuel cells. The first in terms of fuel cell power generation efficiency i Polynomial fitting coefficients of order 1 n To find the highest order of the fitted polynomial, For the load rate of the fuel cell, This is the reference energy efficiency of the fuel cell under rated operating conditions. Long-term performance degradation correction factor for fuel cells ,in, This serves as a correction factor for the long-term performance degradation of fuel cells. For the maintenance level of fuel cells, For fuel cells in the first Annual maintenance level is Voltage attenuation loss at that time This is the rated voltage of the fuel cell at the initial moment.

[0011] More preferably, Short-term variable operating condition correction factor for proton exchange membrane electrolyzer ,in, This is a short-term variable operating condition correction factor for proton exchange membrane electrolyzers. The first percentage of hydrogen production efficiency in proton exchange membrane electrolyzers i Polynomial fitting coefficients of order 1 n To find the highest order of the fitted polynomial, The loading rate of the proton exchange membrane electrolyzer. This is the reference energy efficiency of a proton exchange membrane electrolyzer under rated operating conditions. Long-term performance degradation correction factor for proton exchange membrane electrolyzers , This is a correction factor for the long-term performance degradation of proton exchange membrane electrolyzers. For the maintenance level of the proton exchange membrane electrolyzer, For the proton exchange membrane electrolyzer in the first Annual maintenance level is The irreversible voltage increment at time, This is the rated voltage at the initial moment of the proton exchange membrane electrolyzer.

[0012] More preferably, Short-term variable operating condition correction factor for refrigeration units ,in, This is a short-term variable operating condition correction factor for refrigeration units. This refers to the partial load rate of the refrigeration unit. , , These are the polynomial fitting coefficients, obtained from regression analysis of unit performance test data; Long-term performance degradation correction factor for refrigeration units ,in, This is a correction factor for the long-term performance degradation of the refrigeration unit. For the refrigeration unit at a maintenance level of The annual degradation rate of the performance coefficient, where y is the number of years of operation.

[0013] More preferably, The system energy conversion coupling model is as follows:

[0014] in, This is the output power vector; For the input power vector; This is the equipment dynamic efficiency coupling matrix, where each element represents the equipment dynamic efficiency of each device. For the number of years of operation, , These are the load rate and maintenance level of each device; The charge / discharge coefficient matrix for energy storage devices; This represents the actual charging and discharging power vector of the energy storage device.

[0015] More preferably, In the two-level optimization model, the total lifecycle cost of the system is the sum of the initial investment cost, operating cost, and maintenance cost; wherein, the initial investment cost also includes replacement cost, and the initial investment cost is determined in the following manner:

[0016] in, This refers to the initial investment cost of the system. For equipment n The initial unit investment price; For equipment n Configuration capacity; The total number of devices configured for the system; This represents the total number of equipment replacements during the planning period. For the occurrence of the first k The specific year of the subsequent replacement; For equipment n The unit replacement price; r This is the discount rate for the project.

[0017] More preferably, The total number of replacements of the equipment during the planning period and the occurrence of the [number]th [event / death]. k The specific year of the replacement is determined based on the equipment's maintenance level, specifically:

[0018]

[0019] Where Y represents the system planning period; For equipment n At the set maintenance level The actual usable lifespan is as follows; This indicates rounding down to the nearest integer.

[0020] More preferably, The maintenance cost is determined based on the maintenance level of each piece of equipment, specifically as follows:

[0021] in, To maintain costs; This represents the marginal maintenance cost coefficient. For equipment n The level of maintenance; For equipment n The marginal maintenance cost amplification factor; Maintenance costs to keep the equipment in basic operation; equipment n Configuration capacity.

[0022] More preferably, The two-layer optimization model divides the load rate range of the equipment into multiple linear segments during the solution process of the planning layer. A segmentation identifier variable is introduced in each linear segment, and the nonlinear efficiency curve of the equipment is approximated by a linear equation within each linear segment, thus transforming the nonlinear relationship between the output power and input power of the equipment into a piecewise linear relationship.

[0023] More preferably, Solving the two-layer optimization model specifically includes the following steps: An optimization algorithm is used to generate an initial capacity configuration and maintenance level scheme, and the scheme is then passed to the runtime layer. The execution of the operation layer calculation includes: calculating the long-term performance degradation correction factor of the equipment based on the maintenance level, correcting the piecewise linearization model parameters, and selecting typical year and typical day scenarios to solve the system operation cost under each scenario, and obtaining the full life cycle operation cost by weighted summation; The implementation plan update includes: feeding back the full lifecycle operating cost to the planning layer, and updating the capacity configuration and maintenance level plan in combination with the initial investment cost, maintenance cost and operating cost of the system; The calculation and scheme update process of the runtime layer are executed alternately until the preset convergence condition is met, and the final capacity configuration and maintenance level scheme is output.

[0024] Secondly, the present invention discloses a two-layer optimization system for a comprehensive energy system that considers equipment performance degradation and operation and maintenance based on the aforementioned method, including an equipment dynamic efficiency model construction module, a system energy conversion coupling model construction module, a two-layer optimization model construction module, and a two-layer optimization model solving module; The equipment dynamic efficiency model construction module establishes a dynamic efficiency model for each energy conversion device in an integrated energy system, which couples short-term variable operating condition correction factors with long-term performance degradation correction factors regulated by maintenance level. The system energy conversion coupling model construction module constructs a system energy conversion coupling model based on the equipment dynamic efficiency model of each energy conversion device. The dual-layer optimization model construction module constructs a dual-layer optimization model including a planning layer and an operation layer. The planning layer aims to minimize the total life cycle cost of the system and outputs the configuration capacity and maintenance level of each device. The operation layer aims to minimize the operating cost and constructs constraints based on the system energy conversion coupling model regulated by the configuration capacity and maintenance level. It outputs the operating cost and the output of each device and feeds the operating cost back to the planning layer. The two-level optimization model solving module solves the two-level optimization model to obtain the capacity configuration and operation scheduling scheme of the integrated energy system.

[0025] Thirdly, the present invention provides a terminal, including a processor and a storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of the first aspects of the present invention.

[0026] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects of the present invention.

[0027] The beneficial effects of this invention are compared with those of the prior art: This invention improves the accuracy of life-cycle cost (LCC) estimation and the robustness of planning: Existing technologies mostly use static energy efficiency models, ignoring the cumulative performance degradation of equipment over time, leading to underestimation of operating costs and potential under-planning of capacity. This invention introduces a long-term performance degradation correction factor over a time dimension, accurately depicting the performance evolution trajectory of core equipment such as fuel cells and electrolyzers over a 15-20 year lifecycle, effectively avoiding "prediction bias" and subsequent maintenance risks caused by neglecting aging.

[0028] This invention achieves deep collaborative optimization of "Operation and Maintenance - Planning - Operation": For the first time, "maintenance level" is quantified as a decision variable, and a coupled model of "maintenance investment - degradation rate" is established. This enables the optimization algorithm not only to find the optimal equipment capacity, but also to find the best balance between "high maintenance cost and low degradation" and "low maintenance cost and high degradation". This integrated operation and maintenance perspective fills the gap in traditional IES planning, which only focuses on initial investment and operating costs, while ignoring the impact of operation and maintenance strategies on equipment lifespan.

[0029] This invention resolves the contradiction between nonlinear variable operating conditions and solution efficiency: Addressing the significant nonlinear variable operating conditions (such as efficiency fluctuations under low load) of hydrogen energy cabin equipment (especially electrolyzers and refrigeration units), this invention establishes a refined polynomial fitting model and transforms it into a mixed-integer linear programming (MILP) problem through piecewise linearization. This approach ensures both the fidelity of the physical model and the efficient solution of the complex two-layer model, possessing significant engineering practical value. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the two-layer optimization method for a comprehensive energy system that takes into account equipment performance degradation and operation and maintenance, as described in this invention. Figure 2 This is a diagram of the integrated energy system in Embodiment 1 of the present invention; Figure 3 This is the dynamic energy hub model in Embodiment 1 of the present invention; Figure 4 These are the nonlinear variable operating condition characteristic curves of each key device in Embodiment 1 of the present invention; Figure 5 This is the two-layer optimization model framework in Embodiment 1 of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0032] like Figure 1 As shown, this invention discloses a two-layer optimization method for integrated energy systems that considers equipment performance degradation and operation and maintenance. The method includes the following steps: For each energy conversion device in an integrated energy system, a dynamic efficiency model is established that couples a short-term variable operating condition correction factor with a long-term performance degradation correction factor controlled by maintenance level. The dynamic efficiency of the equipment is the product of the reference energy efficiency of the equipment under rated operating conditions and the short-term variable operating condition correction factor and the long-term performance degradation correction factor; the short-term variable operating condition correction factor is a function of the equipment load rate and is used to represent the dynamic fluctuation characteristics of the equipment efficiency as the load rate changes; the long-term performance degradation correction factor is a function of the equipment's operating years and maintenance level, and the higher the maintenance level, the larger the long-term performance degradation correction factor.

[0033] Preferably, each of the energy conversion devices includes a fuel cell, a proton exchange membrane electrolyzer, and a refrigeration unit; The short-term variable operating condition correction factor for the fuel cell is specifically as follows: ( ,in, This is a short-term variable operating condition correction factor for fuel cells. The first in terms of fuel cell power generation efficiency i Polynomial fitting coefficients of order 1 n To find the highest order of the fitted polynomial, For the load rate of the fuel cell, This is the reference energy efficiency of the fuel cell under rated operating conditions. The long-term performance degradation correction factor for the fuel cell is specifically as follows: ,in, This serves as a correction factor for the long-term performance degradation of fuel cells. For the maintenance level of fuel cells, For fuel cells in the first Annual maintenance level is Voltage attenuation loss at that time This is the rated voltage of the fuel cell at the initial moment; The short-term variable operating condition correction factor for the proton exchange membrane electrolyzer is specifically as follows: ,in, This is a short-term variable operating condition correction factor for proton exchange membrane electrolyzers. The first percentage of hydrogen production efficiency in proton exchange membrane electrolyzers i Polynomial fitting coefficients of order 1 n To find the highest order of the fitted polynomial, The loading rate of the proton exchange membrane electrolyzer. This is the reference energy efficiency of a proton exchange membrane electrolyzer under rated operating conditions. The long-term performance degradation correction factor of the proton exchange membrane electrolyzer is specifically as follows: , This is a correction factor for the long-term performance degradation of proton exchange membrane electrolyzers. For the maintenance level of the proton exchange membrane electrolyzer, For the proton exchange membrane electrolyzer in the first Annual maintenance level is The irreversible voltage increment at time, The rated voltage at the initial moment of the proton exchange membrane electrolyzer; The short-term variable operating condition correction factor for the refrigeration unit is specifically as follows: ,in, This is a short-term variable operating condition correction factor for refrigeration units. This refers to the partial load rate of the refrigeration unit. , , These are the polynomial fitting coefficients, obtained from regression analysis of unit performance test data; The long-term performance degradation correction factor for the refrigeration unit is specifically as follows: ,in, This is a correction factor for the long-term performance degradation of the refrigeration unit. For the refrigeration unit at a maintenance level of The annual degradation rate of the performance coefficient, where y is the number of years of operation.

[0034] Based on the equipment dynamic efficiency model of each energy conversion device, a system energy conversion coupling model is constructed. The system energy conversion coupling model is as follows:

[0035] in, This is the output power vector; For the input power vector; This is the equipment dynamic efficiency coupling matrix, where each element represents the equipment dynamic efficiency of each device. For the number of years of operation, , These are the load rate and maintenance level of each device; The charge / discharge coefficient matrix for energy storage devices; This represents the actual charging and discharging power vector of the energy storage device.

[0036] A two-layer optimization model is constructed, consisting of a planning layer and an operation layer. The planning layer aims to minimize the total lifecycle cost of the system and outputs the configuration capacity and maintenance level of each device. The operation layer aims to minimize the operating cost and constructs constraints based on the system energy conversion coupling model regulated by the configuration capacity and maintenance level. It outputs the operating cost and the output of each device and feeds the operating cost back to the planning layer. In the two-level optimization model, the total lifecycle cost of the system is the sum of the initial investment cost, operating cost, and maintenance cost; wherein, the initial investment cost also includes replacement cost, and the initial investment cost is determined in the following manner:

[0037] in, This refers to the initial investment cost of the system. For equipment n The initial unit investment price; For equipment n Configuration capacity; The total number of devices configured for the system; This represents the total number of equipment replacements during the planning period. For the occurrence of the first kThe specific year of the subsequent replacement; For equipment n The unit replacement price.

[0038] The total number of replacements of the equipment during the planning period and the occurrence of the [number]th [event / death]. k The specific year of the replacement is determined based on the equipment's maintenance level, specifically:

[0039]

[0040] Where Y represents the system planning period; For equipment n At the set maintenance level The actual usable lifespan is as follows; This indicates rounding down to the nearest integer.

[0041] The maintenance cost is determined based on the maintenance level of each piece of equipment, specifically as follows:

[0042] in, To maintain costs; This represents the marginal maintenance cost coefficient. For equipment n The level of maintenance; For equipment n The marginal maintenance cost amplification factor; Maintenance costs to keep the equipment in basic operation; The configuration capacity of device n.

[0043] Solving the two-layer optimization model yields the capacity configuration and operation scheduling scheme of the integrated energy system.

[0044] The two-layer optimization model divides the load rate range of the equipment into multiple linear segments during the solution process of the planning layer. A segmentation identifier variable is introduced in each linear segment, and the nonlinear efficiency curve of the equipment is approximated by a linear equation within each linear segment, thus transforming the nonlinear relationship between the output power and input power of the equipment into a piecewise linear relationship.

[0045] Solving the two-layer optimization model specifically includes the following steps: An optimization algorithm is used to generate an initial capacity configuration and maintenance level scheme, and the scheme is then passed to the runtime layer. The execution of the operation layer calculation includes: calculating the long-term performance degradation correction factor of the equipment based on the maintenance level, correcting the piecewise linearization model parameters, and selecting typical year and typical day scenarios to solve the system operation cost under each scenario, and obtaining the full life cycle operation cost by weighted summation; The implementation plan update includes: feeding back the full lifecycle operating cost to the planning layer, and updating the capacity configuration and maintenance level plan in combination with the initial investment cost, maintenance cost and operating cost of the system; The calculation and scheme update process of the runtime layer are executed alternately until the preset convergence condition is met, and the final capacity configuration and maintenance level scheme is output.

[0046] Example 1: like Figure 1 As shown, this invention discloses a two-layer optimization method for integrated energy systems that considers equipment performance degradation and operation and maintenance. The method includes the following steps: Step 1: For each energy conversion device in the integrated energy system, establish a dynamic efficiency model of the equipment that couples short-term variable operating condition correction factors with long-term performance degradation correction factors regulated by maintenance level. like Figure 2 The diagram shown is a comprehensive energy system diagram in an embodiment of the present invention. The comprehensive energy system is a highly efficient integrated energy system that couples cooling, heating, electricity, and hydrogen. Its physical architecture includes an energy input end, an energy conversion and storage end, and a load end. Energy input sources include wind power, solar power, municipal power grid electricity, and industrial by-product hydrogen.

[0047] The energy conversion and storage end includes a hydrogen-electric-thermal coupling chain and a cold-thermal coupling chain; the hydrogen-electric-thermal coupling chain includes a proton exchange membrane fuel cell (PEMFC), a proton exchange membrane electrolyzer (PEMEC), and a hydrogen storage tank; the cold-thermal coupling chain includes a lithium bromide absorption chiller (utilizing waste heat), an electric chiller, an electric heat pump, a gas boiler, a thermal storage tank, and a cold storage tank.

[0048] The load side includes electrical load, thermal load, cooling load, and hydrogen load.

[0049] The equipment dynamic efficiency model, which couples a short-term variable operating condition correction factor with a long-term performance degradation correction factor controlled by maintenance level, comprehensively considers the equipment's short-term variable operating condition characteristics, long-term performance degradation, and the impact of maintenance level, specifically including: (1) Fuel cell model A. Long-term voltage degradation model considering maintenance levels The output voltage of a fuel cell decreases over time. This invention introduces a maintenance level to correct the rate of degradation. The output voltage can be specifically expressed as:

[0050] in, For fuel cells in the first Annual maintenance level is Output voltage at time: The rated voltage at the initial moment; For fuel cells, the maintenance level must meet the requirements. ; This represents the linear degradation coefficient of the fuel cell due to the level of maintenance. This is a higher-order nonlinear decay term that describes the accelerated decay in the later stages of aging. The nonlinear degradation intensity coefficient can be obtained by fitting long-term aging test data of fuel cells. To run for a period of time y A monotonically increasing nonlinear function, which can be a quadratic function or an exponential function, is used to characterize the accelerated decay effect in the later stages of aging.

[0051] The linear degradation coefficient of the fuel cell affected by maintenance level is modeled as follows:

[0052] in, , These represent the minimum and maximum values ​​of the linear degradation coefficient, respectively, which can be obtained by linear regression calibration of fuel cell voltage degradation data under different maintenance levels; that is, maintenance level. The higher the voltage, the closer the voltage decay rate is to the minimum. Conversely, it approaches the maximum value. .

[0053] B. Short-term variable operating condition efficiency model considering load rate The power generation efficiency and heat generation efficiency of fuel cells are functions of the load rate. A polynomial is used to fit the characteristics of the actual polarization curve:

[0054] in, , These refer to the power generation efficiency and heat generation efficiency of the fuel cell, respectively. The load rate of the fuel cell, and , , They are respectively t The actual output power of the fuel cell during the time period and the rated power (or rated capacity) of the fuel cell. , These are the first three levels of fuel cell power generation efficiency. i The first order polynomial fitting coefficients and the first order of heat generation efficiency i The order polynomial fitting coefficients can be obtained by performing polynomial regression fitting calibration on the measured efficiency data of fuel cells under different load rates. nTo fit the highest order of the polynomial, usually 2 to 4 orders are chosen to balance model accuracy and computational complexity and avoid overfitting caused by higher order terms. This model can accurately reflect the high efficiency in the low-load region (dominated by activation polarization) and the efficiency decline in the high-load region (dominated by ohmic polarization and concentration polarization).

[0055] C. Dynamic efficiency model Combining parts A and B above, the first... y Annual load rate Actual operating power generation efficiency for: =

[0056] in, For fuel cells at a load rate The baseline operating power generation efficiency at that time is the initial variable operating efficiency without taking into account long-term performance degradation. For fuel cells in the first Annual maintenance level is The amount of performance degradation (or voltage attenuation loss) during the process. This is the rated voltage of the fuel cell at the initial moment; This is a short-term variable operating condition correction factor for fuel cells; This is the reference energy efficiency of the fuel cell under rated operating conditions. This is a correction factor for the long-term performance degradation of fuel cells; The short-term variable operating condition correction factor for the fuel cell is specifically as follows: ( ; The long-term performance degradation correction factor for the fuel cell is specifically as follows: .

[0057] This formula shows that long-term degradation will cause the overall efficiency curve of variable operating conditions to shift downward, while a high level of maintenance can mitigate this downward shift.

[0058] (2) Proton exchange membrane electrolyzer model A. Long-term voltage rise model considering maintenance level The performance degradation of a proton exchange membrane electrolyzer is manifested by an increase in the driving voltage over time. Specifically, the voltage increment varies with maintenance level as follows:

[0059]

[0060] in, For irreversible voltage increments, the higher the maintenance level, the slower the voltage rise rate. The maintenance level of the proton exchange membrane electrolyzer is defined by a value ranging from [0,1]. It can be quantified by the ratio of actual maintenance frequency and maintenance quality to the standard maintenance plan. 1 represents the highest maintenance level, and 0 represents zero maintenance. The linear voltage decay rate of a proton exchange membrane electrolyzer affected by maintenance level; For proton exchange membrane electrolyzers at the highest maintenance level (i.e. The minimum linear voltage decay rate at (when = 1), For proton exchange membrane electrolyzers to operate at a zero-maintenance level (i.e. The maximum linear voltage decay rate at (=0) can be obtained by linear regression calibration of long-term aging test data of electrolytic cells under different maintenance levels. y The service life of the equipment; The secondary voltage decay coefficient (characterizing the nonlinear acceleration feature of equipment aging) can be obtained by fitting voltage change data in the later stage of long-term operation of the electrolytic cell.

[0061] B. Short-term variable operating condition model The efficiency of hydrogen production in an electrolyzer exhibits a significant nonlinearity with input power (increasing first and then decreasing), specifically:

[0062] in, for t The actual hydrogen production efficiency of the time-phase electrolyzer; The highest order of the polynomial fitting; The first efficiency of hydrogen production in electrolyzers i The order polynomial fitting coefficients are obtained by fitting the polarization curve of the electrolytic cell at the factory or historical operating data using the least squares method. for t The actual input electrical power of the time-phase electrolytic cell; The rated input power of the electrolytic cell; the ratio term in the formula ( / This means that the electrolytic cell is in t Partload ratio for a given time period.

[0063] C. Dynamic efficiency model Combining parts A and B above, the load rate in year y is... The actual hydrogen production efficiency at that time is:

[0064] in, This is the reference energy efficiency of a proton exchange membrane electrolyzer under rated operating conditions. This is a short-term variable operating condition correction factor for proton exchange membrane electrolyzers; This is a correction factor for the long-term performance degradation of proton exchange membrane electrolyzers. The short-term variable operating condition correction factor for the proton exchange membrane electrolyzer is specifically as follows: ,in, The first percentage of hydrogen production efficiency in proton exchange membrane electrolyzers i Polynomial fitting coefficients of order 1 n To find the highest order of the fitted polynomial, The loading rate of the proton exchange membrane electrolyzer. This is the reference energy efficiency of a proton exchange membrane electrolyzer under rated operating conditions. The long-term performance degradation correction factor of the proton exchange membrane electrolyzer is specifically as follows: ,in, For the maintenance level of the proton exchange membrane electrolyzer, For the proton exchange membrane electrolyzer in the first Annual maintenance level is The irreversible voltage increment at time, This is the rated voltage at the initial moment of the proton exchange membrane electrolyzer.

[0065] (3) Refrigeration unit (absorption and electric refrigeration) model A. Long-term performance degradation model considering maintenance levels The COP of refrigeration units decreases exponentially over time:

[0066] in, For the first y The actual coefficient of performance of the refrigeration unit; This refers to the initial rated performance coefficient of the refrigeration unit; y For the number of years of operation; The annual degradation rate of the coefficient of performance of the refrigeration unit; In this invention, the annual decline rate The function for maintaining the level is defined in its specific functional form as follows:

[0067] in, For the refrigeration unit at a given maintenance level The actual annual recession rate at that time; The maintenance level of the refrigeration unit, and meets the requirements. ∈[0,1]; For refrigeration units with zero maintenance investment ( The maximum annual baseline decline rate when (=0); For the refrigeration unit at the highest maintenance investment ( The minimum annual limit of decline that can be achieved when =1).

[0068] B. Short-term variable operating condition model For absorption chillers, a COP-PLR curve model with an initial increase followed by a decrease is adopted, with peak efficiency set in the 50%-60% load rate range; for electric chillers, an empirical model based on EIR (Energy Input Ratio) is adopted. This invention uses a unified form of quadratic polynomial model to describe the variable operating condition performance of the refrigeration unit. The actual COP expression is:

[0069] in, This refers to the coefficient of performance (COP) of the refrigeration unit under actual operating conditions. The coefficient of performance (COP) of the refrigeration unit under rated operating conditions; This refers to the partial load rate of the refrigeration unit, i.e., the load rate. , , These are the polynomial fitting coefficients, obtained from the regression of unit performance test data.

[0070] C. Dynamic efficiency model Combining parts A and B above, the refrigeration unit in year y has a load rate of The specific cooling efficiency at that time is:

[0071] in, This is a short-term variable operating condition correction factor for refrigeration units; This is a correction factor for the long-term performance degradation of the refrigeration unit. The short-term variable operating condition correction factor for the refrigeration unit is specifically as follows: ; The long-term performance degradation correction factor for the refrigeration unit is specifically as follows: ; (4) Maintenance cost model To establish a trade-off mechanism, the unit maintenance cost of each piece of equipment Modeled as maintenance level Nonlinear increasing function:

[0072] in, This is the marginal maintenance cost coefficient, used to characterize the pursuit of an extremely low decay rate (i.e., This requires a sharp increase in marginal maintenance costs. The preferred value is a real number greater than 2, with a typical value range of [2.0, 4.0]. For equipment nThe marginal maintenance cost amplification factor (or the maximum variable maintenance cost extreme value), the factor Its physical meaning is: the equipment maintenance level is raised to the theoretical limit (i.e., When =1), compared to only maintaining basic operation ( The additional cost cap; its typical value can be expressed by the formula. = - It is confirmed that, among them, This represents the highest single / single-cycle maintenance cost that has occurred or is estimated in historical statistics. The specific absolute value varies depending on the cost of different equipment types.

[0073] and These are empirical fitting parameters. Their actual values ​​are obtained by collecting historical operation and maintenance records of specific equipment (such as electrolyzers, fuel cells, or refrigerators), full life-cycle maintenance manuals provided by equipment manufacturers, and reliability test data. The actual maintenance cost is used as the vertical axis and the corresponding maintenance level index is used as the horizontal axis. The parameters are obtained by using the least squares method to fit a nonlinear curve.

[0074] Step 3: Based on the equipment dynamic efficiency model of each energy conversion device, construct a system energy conversion coupling model; Traditional energy hub (EH) models typically employ static transformation matrices, which fail to reflect the time-varying characteristics of device performance. This invention proposes a dynamically modified energy conversion coupling model that maps the dual dynamic characteristics of individual devices to the system level.

[0075] (1) Definition of dynamic efficiency Any energy conversion path within the system ( Actual conversion efficiency Defined as the product of three factors:

[0076] in, This is a short-term variable operating condition correction factor. For equipment load rate, It is a nonlinear function (such as a polynomial or piecewise function) that describes the performance fluctuation of the equipment under partial load; It is a long-term recession correction factor. y For the number of years of operation, The maintenance level of the equipment describes the decline in the performance baseline caused by the aging of the equipment over time, and the rate of decline is controlled by the maintenance level. This is the reference energy efficiency (design parameters) for the new equipment under rated operating conditions.

[0077] (2) Dynamic transformation matrix Based on the above definition, a modified energy conversion coupling model is constructed:

[0078] in, Output power vector (electric, thermal, cold, hydrogen load); For input power vectors (grid, gas grid, renewable energy); This is the equipment dynamic efficiency coupling matrix, where each element represents the equipment dynamic efficiency of each device. For the number of years of operation, , The load rate and maintenance level of each device are respectively, and their specific calculation method is directly taken from the dynamic actual conversion efficiency defined in step (1). This means that the matrix The element values ​​corresponding to different energy conversion paths are no longer constants in the traditional EH model, but instead include the equipment aging state. Real-time load rate and maintenance level Dynamic time-varying characteristics; The energy storage device's charge and discharge coefficient matrix is ​​a constant mapping matrix. The matrix elements reflect the topology of each energy storage device connected to the internal bus (electricity, heat, hydrogen) of the system, as well as the fixed loss coefficient during the charge and discharge process (if the loss is not calculated separately in the SOC). This refers to the actual charging and discharging power vector of the energy storage device. In the symbol definition system of this invention, a vector... If the element value in is positive (E) i A value >0 indicates that the energy storage device is charging / storing energy (i.e., absorbing energy as an internal load of the system); a negative value (E) .... i <0 indicates that the energy storage device is discharging / releasing energy (i.e., releasing energy as the internal power source of the system).

[0079] Specifically, the total power ultimately supplied by the system to the external load It should be equal to the total input conversion power provided by each conversion device. Subtract the power absorbed by the internal energy storage device. When the energy storage is charging ( (positive) Subtracting this positive value indicates that some of the converted energy flows to energy storage, reducing the available energy for the load; when the energy storage discharges ( (Negative) indicates that the energy released by the energy storage is added to the energy of the conversion equipment to jointly meet the external load demand.

[0080] Specifically, such as Figure 3The diagram illustrates the dynamic energy hub theoretical model constructed in this invention, which considers the varying operating conditions of equipment. Unlike traditional static energy hub models that use a constant transformation matrix, this invention introduces dynamic nonlinear efficiency, achieving a precise mapping between the underlying physical state of the equipment and the macroscopic energy distribution of the system; wherein, the input power port... The diagram represents the input power of different types of energy received by the integrated energy system at the source side (e.g., grid power, natural gas pipeline power, external hydrogen source input, etc.). Different line styles (solid lines, dashed lines, dotted lines) in the diagram represent independent energy flows with different physical properties within the system (e.g., electrical flow, heat flow, hydrogen flow); terminal load port. This represents the different types of terminal load demands that the system needs to meet on the load side (e.g., user electrical load, thermal load, hydrogen load); dynamic conversion efficiency module. This represents the energy conversion efficiency of various energy conversion devices within the system (such as fuel cells, electrolyzers, and refrigeration units). For example, Indicates that the energy carrier will be input. Converted into an energy output carrier Actual efficiency at that time; energy storage device module ( ) represents a device mounted on a specific energy bus (such as Internal energy storage units (such as hydrogen storage tanks, battery banks, or thermal / cold storage tanks) on nodes. Within the energy hub architecture, it acts as an energy buffer to smooth power fluctuations at both the source and load ends; as shown by the dotted dashed lines (feedback loops) in the diagram, the load state (N) at the system terminal is fed down into the comprehensive correction equation, which calculates in real time the result that integrates micro-level variable operating conditions and macro-level long-term degradation (such as...). Figure 4 (Short-term variable operating condition characteristic curves of each device shown) Real-time dynamic efficiency of dual physical characteristics The correction signal is transmitted from bottom to top to the efficiency module of each physical device via a bidirectional arrow through the control bus on the left. This architecture achieves deep mathematical decoupling and precise physical mapping between device degradation status (upper-level decision) and intraday load allocation (lower-level decision).

[0081] Step 4: Construct a two-layer optimization model including a planning layer and an operation layer; wherein, the planning layer aims to minimize the total life cycle cost of the system and outputs the configuration capacity and maintenance level of each device, and the operation layer aims to minimize the operating cost. Based on the system energy conversion coupling model regulated by the configuration capacity and maintenance level, it constructs constraints, outputs the output and operating cost of each device, and feeds the operating cost back to the planning layer. To address the time scale differences between long-term planning (e.g., 15-20 years) and short-term operation (e.g., hourly), as well as the coupling relationship between equipment degradation and maintenance strategies, this invention constructs a two-layer "planning-operation" optimization model, such as... Figure 5 As shown.

[0082] A. Planning layer, i.e., upper-level model: used to decide on the configuration capacity and maintenance level of each device throughout its entire life cycle; In the planning layer, decision variables include the equipment configuration capacity and equipment maintenance level of each device; The equipment configuration capacity of each device includes, but is not limited to, the rated capacity of fuel cells, electrolyzers, refrigeration units, and energy storage devices.

[0083] For critical, easily degraded equipment, including fuel cells, electrolyzers, and refrigerators, establish maintenance levels throughout their entire lifecycle. .

[0084] The maintenance level It is the pivot in this invention that connects the underlying physical aging laws of the system with the upper-level economic decision-making costs. The numerical value intuitively reflects the "depth and frequency" of preventive and predictive maintenance. When it approaches 0, it indicates that only a basic "reactive maintenance" strategy is adopted; when... When the value approaches 1, it indicates that extremely frequent and in-depth maintenance operations have been performed (such as periodic optimization of membrane electrode hydrothermal management, catalyst activity restoration, or early replacement of electrolyzer seals for fuel cells).

[0085] The objective function of the planning layer is to minimize the total life cycle cost (LCC), specifically:

[0086] in: The total cost of the system's entire lifecycle; This refers to the initial investment cost of the system. Let y be the annual operating cost in year y. Let y be the annual maintenance cost in year y, where y is the operating lifespan, r is the project's discount rate used to convert future costs into present value, and Y is the system planning period, i.e., the total lifespan.

[0087] The initial investment cost of the system depends on the unit price of the equipment and the configuration capacity. If the equipment is scrapped prematurely within the planning period Y due to low maintenance level (efficiency below the threshold), the equipment replacement cost must be included, specifically:

[0088] in, Let n be the unit initial investment price of equipment n; Let n be the configured capacity of device n; The total number of devices configured for the system; This represents the total number of equipment replacements during the planning period. For the occurrence of the first k The specific year of the subsequent replacement; Let n be the unit replacement price for equipment. Considering technological advancements, the replacement price may be equal to or lower than the initial price. The actual service life of equipment is affected by the level of maintenance. Dynamic control means that the lower the maintenance level, the faster the degradation and the shorter the lifespan. Therefore, the total number of replacements of equipment within the planning period... Passing the exam The year in which the replacement occurred The calculation is as follows:

[0089]

[0090] in, For equipment n at the set maintenance level The actual usable lifespan is as follows; This indicates the floor function.

[0091] The annual operating cost of the system is derived from the operating layer and is affected by the aging status of the equipment in that year. The more severe the aging, the lower the energy efficiency and the greater the fuel / electricity consumption.

[0092] Annual maintenance costs It is directly linked to the level of maintenance, specifically:

[0093] in, The maintenance cost in year y; In this embodiment, the marginal maintenance cost coefficient is used. The preferred value is a real number greater than 2, with a typical value range of [2.0, 4.0]. For equipment n The level of maintenance; To maintain equipment n Basic operating maintenance costs; For equipment n The marginal maintenance cost amplification factor, whose typical value can be obtained through the formula... = - Sure, This represents the highest single / single-cycle maintenance cost that has occurred or is estimated in historical statistics. The specific absolute value varies depending on the cost of different equipment types. The configuration capacity of device n. This reflects the economic cost of "high investment in exchange for low recession".

[0094] The constraints at the planning level include: (1) Equipment capacity constraints: ,in, equipment n Configuration capacity, , The first n The minimum and maximum configuration capacity of each device; (2) System reliability constraints, including load failure rate, system reserve capacity, etc., specifically: The load failure rate constraint is used to ensure that the percentage of energy that cannot meet user load demands during the entire planning period is below a set safety threshold.

[0095] in, For the first y Year t The actual power loss due to insufficient capacity, equipment degradation, or extreme weather in the system at any time (i.e., the power / heat / hydrogen load shortfall that the system cannot meet, which is fed back by the lower-level scheduling model). For the first y Year t The total predicted load demand of the system at any given time; Y and T represent the planning period and the total number of scheduling periods within the year, respectively. This indicator represents the maximum allowable load failure rate threshold for the system (typically 1% to 5%), and it rigidly defines the reliability baseline for system planning.

[0096] System reserve capacity constraints are used to constrain the total stable and schedulable capacity configured in the system. This capacity must be able to cover extreme peak loads and provide a certain margin.

[0097] in, This refers to a collection of energy supply and storage devices with stable and dispatchable capabilities within the system (such as fuel cells, refrigeration units, energy storage power stations, etc., excluding wind and solar equipment with strong randomness). equipment n Configuration capacity; Equipment whose maximum usable capacity is reduced due to aging as defined above. n Long-term performance degradation correction factor; For the first y The maximum peak load predicted for the year; This is the system's reserve capacity margin factor (e.g., 1%~5%).

[0098] (3) Maintenance level constraints: .

[0099] B. Runtime layer, i.e., the lower-level model, is used for intraday optimization and scheduling in multiple scenarios; The input parameters of the operating layer include the device capacity passed from the upper layer. Maintenance level And calculate the year y (which determines the current long-term recession correction factor).

[0100] The decision variables of the operating layer include the power output of each device at typical times of the day. And the energy storage charging and discharging status.

[0101] The objective function of the runtime layer, used to minimize daily operating costs, is as follows:

[0102] in, Daily operating costs, , , The first t The amount of electricity, natural gas, and hydrogen purchased during each period. , , The first t Electricity purchase price, natural gas purchase price, and hydrogen purchase price for each time period; For the system in the first t Revenue generated from selling energy (such as electricity, cooling, heat, or hydrogen) to external parties during a given period.

[0103] Constraints: (1) Power balance constraint and dynamic efficiency constraint: that is, the real-time supply and demand balance of electricity, heat, cold and hydrogen including dynamic efficiency, which is calculated by the energy conversion coupling model in step 3, and the matrix in the energy conversion coupling model. This reflects dynamic efficiency constraints, while matrix operation relationships reflect power balance constraints of multi-energy flows; When equipment ages prematurely or operates within a suboptimal load range, the dynamic matrix... The efficiency element in the vector becomes smaller, which means that the rigid external load demand vector must be met. Under the power balance constraint, the system is inevitably forced to increase the amount of external energy purchased, which increases the product of the terms in the objective function and directly leads to a higher total daily operating cost. Significant increase; (2) Equipment output constraints: and the slope rate constraint ;in, , respectively equipment n exist t +1 time period t Actual output during the time period For equipment nMaximum capacity; , respectively equipment n The minimum and maximum gradeability; (3) Energy storage constraints: State of charge (SOC) continuity constraint and initial and final SOC balance constraint.

[0104] In the two-layer optimization model of this invention, the device capacity passed from the upper-layer model is... The upper-level maintenance level and the calculation year together determine the degree of degradation and downward shift of the efficiency curve used by the lower-level model during invocation (i.e., the long-term performance degradation correction factor). Based on this dynamic efficiency curve discounted by the upper-level state, the lower-level model performs precise intraday optimization to obtain the minimum intraday operating cost. This information is then passed back to the upper level and aggregated into annual operating costs. This forms an economic feedback mechanism for evaluating the total life cycle cost (LCC) at the upper level, achieving synergy between "configuration-maintenance-operation".

[0105] Step 4: Solve the two-layer optimization model to obtain the capacity configuration and operation scheduling scheme of the integrated energy system.

[0106] The model solution strategy involves solving the problem through a combination of linearization and algorithmic interaction. This two-layer model contains nonlinear efficiency curves (variable operating condition characteristics) and nonlinear coupling relationships (aging and maintenance), making it difficult to solve. This invention adopts the following strategy: Linearization of nonlinear constraints: For the nonlinear variable operating condition efficiency curve of equipment in the operating layer A piecewise linearization method is used, where For input power, This refers to the output power. Divide the equipment load rate range [0, 1] into... K One linear segment; Introducing binary variables Identification equipment at all times t Running in k On each segment; Within each segment, use linear equations Approximate original curve, where , For the first k Linearized model parameters for each segment.

[0107] Through this process, the lower-level optimization problem is transformed into a mixed-integer linear programming (MILP) problem, which can be efficiently solved using mature commercial solvers.

[0108] Two-layer algorithm interaction process: Step 4.1: Use an optimization algorithm to generate an initial capacity configuration and maintenance level scheme; In this embodiment, preferably, an improved genetic algorithm (GA) is used as the upper-level solver to initialize the population and generate a set of capacity configuration and maintenance level schemes; Step 4.2: Pass the scheme to the operation layer. The operation layer calculates the long-term performance degradation correction factor of the equipment for each year based on the maintenance level, and corrects the piecewise linearization parameters in the operation layer model. Step 4.3: In the runtime layer, select typical year and typical day scenarios to solve the system runtime cost under each scenario, and obtain the full life cycle runtime cost by weighted summation; Specifically, for typical years (such as years 1, 5, 10, and 15) and typical days (spring, summer, autumn, and winter) throughout the entire lifecycle, solve the lower-level MILP problems corresponding to each typical year and typical day to obtain the optimal operating cost for each scenario. The total lifecycle operating cost is obtained by weighted summation; Step 4.4: Feed back the total lifecycle operating cost to the planning layer, calculate the fitness value (i.e., the total lifecycle cost) by combining the initial investment cost and maintenance cost of the system, and update the capacity configuration and maintenance level scheme; Specifically, the planning layer GA performs selection, crossover, and mutation to update the population and generate new capacity configuration and maintenance level schemes; Step 4.5: Alternately execute steps 4.2-4.4 until the preset convergence condition is met, and output the final capacity configuration and maintenance level scheme.

[0109] The preset convergence condition is: the number of iterations of the optimization algorithm reaches the upper limit, or the rate of change of the total cost over the entire life cycle is lower than the preset rate of change threshold in multiple consecutive iterations.

[0110] This invention constructs a dynamic equipment model that couples "short-term variable operating conditions + long-term degradation + maintenance level" and introduces maintenance level as a decision variable in the two-layer optimization architecture. This enables the coordinated optimization of capacity configuration, operation scheduling and maintenance strategies throughout the entire life cycle of the integrated electric-hydrogen-thermal energy system, thereby achieving the goal of minimizing the overall system cost and maximizing reliability.

[0111] Example 2: To more clearly illustrate the outstanding substantive features of this invention and the significant progress it brings to the prior art, an application example of implementing this invention is described below.

[0112] To verify the effectiveness of the two-layer optimization method for integrated energy systems that considers equipment performance degradation and operation and maintenance proposed in this invention, this embodiment sets up four typical comparative scenarios for simulation analysis, and the system planning period is set to 15 years.

[0113] Specifically, the following scenarios are set up: Scenario 1 (No Degradation, Ideal Control Group): Assuming that all devices maintain constant performance throughout their entire lifecycle, with no degradation (ideal model).

[0114] Scenario 2 (Mild Degradation): The equipment degradation rate is relatively slow. The equipment performance degradation factors for years 0, 5, 10, and 15 are [1, 0.95, 0.9, 0.85], representing high maintenance levels or high-quality equipment scenarios.

[0115] Scenario 3 (Moderate Degradation): Considering the linear degradation of equipment performance over years of operation. The equipment performance degradation factors for years 0, 5, 10, and 15 are [1, 0.95, 0.9, 0.85], representing scenarios with moderate maintenance levels or moderate quality equipment.

[0116] Scenario 4 (Severe Degradation): The equipment degrades at a relatively fast rate. The equipment performance degradation factors for years 0, 5, 10, and 15 are [1, 0.875, 0.75, 0.625], representing low maintenance levels or harsh operating conditions.

[0117] 2. Analysis of Optimized Configuration Results Table 1 shows the optimal equipment capacity configuration (unit: kW) and total lifecycle cost (LCC) of the system under four scenarios.

[0118] Table 1 Comparison of Equipment Capacity Configuration and Total Cost in Different Scenarios

[0119] Results analysis: The effect of adjusting the life cycle cost (LCC): Comparing Scenario 1 with Scenarios 2-4 reveals that the traditional model, which ignores degradation, severely underestimates system costs. The LCC for Scenario 1 is only 282 million yuan, while considering actual degradation, even a mild degradation (Scenario 2) costs 452 million yuan, and a severe degradation (Scenario 4) costs a staggering 2.135 billion yuan. This demonstrates that ignoring equipment aging can lead to significant deviations in investment budgets (up to seven times or more). The method of this invention more accurately reflects the economic risks of a project.

[0120] Impact on hydrogen energy system capacity planning: Observing the configuration of electrolyzers (Ely) and hydrogen storage tanks reveals a significant trend: In the case of mild degradation (Scenario 2), the system tends to be configured with a larger electrolyzer (1774kW) and hydrogen storage tank (952kW) to combat the later efficiency decline and ensure a continuous hydrogen supply.

[0121] However, under severe degradation (Scenario 4), the electrolyzer capacity was significantly reduced to 480kW. This is because maintaining a large self-made hydrogen system becomes economically unfeasible when equipment degrades too rapidly (maintenance and efficiency loss costs are too high). The optimization algorithm automatically selected a strategy of reducing the initial capacity and relying on external energy sources, avoiding over-investment in rapidly aging equipment.

[0122] Furthermore, a comparison of typical annual operating costs is conducted, as follows: To deeply analyze the impact of equipment performance degradation on the economics of the system throughout its entire lifecycle, the total annual operating costs (including energy purchase costs, operation and maintenance costs, and off-load penalty costs) for four scenarios were extracted in years 1, 5, 10, and 15. Data comparisons are shown in Table 2.

[0123] Table 2 Comparison of typical annual total operating costs under four scenarios (unit: RMB 10,000)

[0124] Based on the data in Table 2, the following key conclusions can be drawn: The static nature of the ideal model and its deviation from reality (Scenario 1): Scenario 1 assumes that the equipment performance is constant, and its annual operating cost remains stable at 17.74 million yuan. This figure, used as a baseline, masks the significant risks that the system may face in the later stages.

[0125] The "avalanche effect" of recession (Scenario 3 and Scenario 4): In moderate recession (Scenario 3) and severe recession (Scenario 4), operating costs exhibit an exponential growth trend. Taking Scenario 4 as an example, the operating cost in the 15th year reached 308 million yuan, which is more than 14 times that of the first year.

[0126] Cause Analysis: As the operating years increase, the efficiency factors of fuel cells and electrolyzers decrease significantly (dropping to 0.625 in year 15 of Scenario 4). This not only increases the raw material consumption per unit of energy output (leading to a surge in grid power purchases, see previous analysis), but more seriously, due to initial investment constraints during capacity planning, aging equipment cannot meet peak load demands in later stages, resulting in high penalty costs for load shedding. For example, in year 15 of Scenario 4, the vast majority of operating costs stem from penalties for insufficient cold / heat / hydrogen load. This clearly demonstrates that without considering degradation, the system will face a severe power supply reliability crisis at the end of its lifespan.

[0127] The economic benefits of a "high-end anti-aging" strategy (Scenario 2): The cost of Scenario 2 (mild degradation / high maintenance) in the first year (RMB 13.42 million) was significantly lower than that of Scenario 1 (RMB 17.74 million) without degradation. This is due to the optimized configuration of the electrolyzer and energy storage capacity in this scenario (see Table 1: electrolyzer 1774kW vs. Scenario 1 1485kW; hydrogen storage tank 952kW vs. Scenario 1 685kWh; thermal storage 367kW vs. Scenario 1 0kWh; cold storage 997kW vs. Scenario 1 581kWh). The high-efficiency equipment and sufficient capacity in the initial stage enabled the system to significantly reduce the cost of purchased energy through "self-generation and self-consumption".

[0128] Even in the 15th year, the operating cost of Scenario 2 (71.39 million yuan) is only about one-quarter of that of Scenario 4 (308 million yuan). This shows that by improving maintenance levels (mitigating decline) or reserving capacity margins during the planning period, although the initial investment is increased, the runaway operating costs in the later stages can be significantly curbed, achieving the optimal total life cycle cost (LCC) (Scenario 2 LCC 452 million yuan << Scenario 4 LCC 2.135 billion yuan).

[0129] Changes in grid interaction costs In addition to the total cost, the changes in the cost of directly purchasing electricity also confirm the logic of "equipment aging - reliance on external grid".

[0130] In Scenario 4, the cost of purchasing electricity from the grid soared from 750,000 yuan in the first year to 3.65 million yuan in the 15th year (an increase of about 5 times).

[0131] This indicates that as the performance of self-generating equipment (fuel cells) declines, the system is forced to transform from a "producer" into a mere "consumer," a passive role shift that traditional static programming methods could not foresee.

[0132] This embodiment demonstrates the use of the integrated operation and maintenance optimization method proposed in this invention: It avoids investment misjudgments: it identifies a serious underestimation of LCC by the static model (error range 60%-650%).

[0133] The capacity structure has been optimized: based on the characteristics of equipment degradation, the system intelligently balances the two paths of "over-investing in the initial stage to resist degradation" and "increasing external energy purchases in the later stage", avoiding blind expansion of easily degradable equipment (such as electrolyzers).

[0134] Achieving full-cycle optimization: Through the bidirectional coupling of the planning layer and the operation layer, the system ensures that it maintains the optimal balance between economy and reliability throughout the entire process of equipment from "youth" to "old age".

[0135] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0136] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0137] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0138] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A two-layer optimization method for a comprehensive energy system considering equipment performance degradation and operation and maintenance, characterized in that, The method includes the following steps: For each energy conversion device in an integrated energy system, a dynamic efficiency model is established that couples a short-term variable operating condition correction factor with a long-term performance degradation correction factor controlled by maintenance level. The dynamic efficiency of the device is the product of the reference energy efficiency of the device under rated operating conditions and the short-term variable operating condition correction factor and the long-term performance degradation correction factor. The short-term variable operating condition correction factor is a function of the device load rate and is used to represent the dynamic fluctuation characteristics of the device efficiency as the load rate changes. The long-term performance degradation correction factor is a function of the device's service life and maintenance level; the higher the maintenance level, the larger the long-term performance degradation correction factor. Based on the equipment dynamic efficiency model of each energy conversion device, a system energy conversion coupling model is constructed. A two-layer optimization model is constructed, comprising a planning layer and an operation layer. The planning layer aims to minimize the system's total lifecycle cost, outputting the configuration capacity and maintenance level of each device. The operation layer aims to minimize operating costs, constructing constraints based on the system energy conversion coupling model regulated by configuration capacity and maintenance level, outputting the operating cost and the output of each device, and feeding the operating cost back to the planning layer. The total system lifecycle cost is the sum of the initial investment cost, operating cost, and maintenance cost. The initial investment cost also includes replacement costs, which are determined as follows: ,in, This refers to the initial investment cost of the system; For equipment n The initial unit investment price; For equipment n Configuration capacity; The total number of devices configured for the system; This represents the total number of equipment replacements during the planning period. For the occurrence of the first k The specific year of the replacement is determined based on the equipment's maintenance level. For equipment n The unit replacement price; r The discount rate for the project; The maintenance cost is determined based on the maintenance level of each piece of equipment, specifically as follows: ,in, To maintain costs; This represents the marginal maintenance cost coefficient. For equipment n The level of maintenance; For equipment n The marginal maintenance cost amplification factor; Maintenance costs to keep the equipment in basic operation; equipment n Configuration capacity; Solving the two-layer optimization model yields the capacity configuration and operation scheduling scheme of the integrated energy system.

2. The two-layer optimization method for integrated energy systems according to claim 1, characterized in that: The energy conversion devices include fuel cells, proton exchange membrane electrolyzers, and refrigeration units.

3. The two-layer optimization method for integrated energy systems according to claim 2, characterized in that: Short-term variable operating condition correction factor for fuel cells ( ,in, This is a short-term variable operating condition correction factor for fuel cells. The first in terms of fuel cell power generation efficiency i Polynomial fitting coefficients of order 1 n To be the highest order of the fitted polynomial, For the load rate of the fuel cell, This is the reference energy efficiency of the fuel cell under rated operating conditions. Long-term performance degradation correction factor for fuel cells ,in, This serves as a correction factor for the long-term performance degradation of fuel cells. For the maintenance level of fuel cells, For fuel cells in the first Annual maintenance level is Voltage attenuation loss at that time This is the rated voltage of the fuel cell at the initial moment.

4. The two-layer optimization method for integrated energy systems according to claim 2, characterized in that: Short-term variable operating condition correction factor for proton exchange membrane electrolyzer ,in, This is a short-term variable operating condition correction factor for proton exchange membrane electrolyzers. The first percentage of hydrogen production efficiency in proton exchange membrane electrolyzers i Polynomial fitting coefficients of order 1 n To be the highest order of the fitted polynomial, The loading rate of the proton exchange membrane electrolyzer. This is the reference energy efficiency of a proton exchange membrane electrolyzer under rated operating conditions. Long-term performance degradation correction factor for proton exchange membrane electrolyzers ,in, This is a correction factor for the long-term performance degradation of proton exchange membrane electrolyzers. For the maintenance level of the proton exchange membrane electrolyzer, For the proton exchange membrane electrolyzer in the first Annual maintenance level is The irreversible voltage increment at time, This is the rated voltage at the initial moment of the proton exchange membrane electrolyzer.

5. The two-layer optimization method for integrated energy systems according to claim 2, characterized in that: Short-term variable operating condition correction factor for refrigeration units ,in, This is a short-term variable operating condition correction factor for refrigeration units. This refers to the partial load rate of the refrigeration unit. , , These are the polynomial fitting coefficients, obtained from regression analysis of unit performance test data; Long-term performance degradation correction factor for refrigeration units ,in, This is a correction factor for the long-term performance degradation of the refrigeration unit. For the refrigeration unit at a maintenance level of The annual degradation rate of the performance coefficient, where y is the number of years of operation.

6. The two-layer optimization method for integrated energy systems according to claim 1, characterized in that: The system energy conversion coupling model is as follows: in, This is the output power vector; For the input power vector; This is the equipment dynamic efficiency coupling matrix, where each element represents the equipment dynamic efficiency of each device. For the number of years of operation, , These are the load rate and maintenance level of each device; The charge / discharge coefficient matrix for energy storage devices; This represents the actual charging and discharging power vector of the energy storage device.

7. The two-layer optimization method for integrated energy systems according to claim 6, characterized in that: The total number of replacements of the equipment during the planning period and the occurrence of the [number]th [event / death]. k The specific year of the replacement is as follows: Where Y represents the system planning period; For equipment n At the set maintenance level The actual usable lifespan is as follows; This indicates rounding down to the nearest integer.

8. The two-layer optimization method for integrated energy systems according to claim 1, characterized in that: The two-layer optimization model divides the load rate range of the equipment into multiple linear segments during the solution process of the planning layer. A segmentation identifier variable is introduced in each linear segment, and the nonlinear efficiency curve of the equipment is approximated by a linear equation within each linear segment, thus transforming the nonlinear relationship between the output power and input power of the equipment into a piecewise linear relationship.

9. The two-layer optimization method for integrated energy systems according to claim 8, characterized in that: Solving the two-layer optimization model specifically includes the following steps: An optimization algorithm is used to generate an initial capacity configuration and maintenance level scheme, and the scheme is then passed to the runtime layer. The execution of the operation layer calculation includes: calculating the long-term performance degradation correction factor of the equipment based on the maintenance level, correcting the piecewise linearization model parameters, and selecting typical year and typical day scenarios to solve the system operation cost under each scenario, and obtaining the full life cycle operation cost by weighted summation; The implementation plan update includes: feeding back the full lifecycle operating cost to the planning layer, and updating the capacity configuration and maintenance level plan in combination with the initial investment cost, maintenance cost and operating cost of the system; The calculation and scheme update process of the runtime layer are executed alternately until the preset convergence condition is met, and the final capacity configuration and maintenance level scheme is output.

10. A two-layer optimization system for a comprehensive energy system considering equipment performance degradation and operation and maintenance, based on the method of any one of claims 1-9, comprising an equipment dynamic efficiency model construction module, a system energy conversion coupling model construction module, a two-layer optimization model construction module, and a two-layer optimization model solving module, characterized in that: The equipment dynamic efficiency model construction module establishes a dynamic efficiency model for each energy conversion device in an integrated energy system, which couples short-term variable operating condition correction factors with long-term performance degradation correction factors regulated by maintenance level. The system energy conversion coupling model construction module constructs a system energy conversion coupling model based on the equipment dynamic efficiency model of each energy conversion device. The dual-layer optimization model construction module constructs a dual-layer optimization model including a planning layer and an operation layer. The planning layer aims to minimize the total life cycle cost of the system and outputs the configuration capacity and maintenance level of each device. The operation layer aims to minimize the operating cost and constructs constraints based on the system energy conversion coupling model regulated by the configuration capacity and maintenance level. It outputs the operating cost and the output of each device and feeds the operating cost back to the planning layer. The two-level optimization model solving module solves the two-level optimization model to obtain the capacity configuration and operation scheduling scheme of the integrated energy system.

11. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-9.

Images