PEM electrolytic cell energy-economy collaborative optimization method based on digital twinning

Abstract

The invention relates to the technical field of electrolytic hydrogen production, and discloses a PEM electrolytic cell energy-economy collaborative optimization method based on digital twinning, which comprises the following steps: S1, calibrating parameters of a membrane electrode mathematical model by using a standard test for a single cell of a PEM electrolytic cell, and establishing a high-precision physical simulation model; s2, constructing a high-precision digital twin system based on the high-precision physical simulation model; s3, inputting multi-dimensional data of electricity price fluctuation, hydrogen demand prediction and equipment state, and performing collaborative optimization analysis through an optimization decision algorithm; s4, forming a PEM electrolytic cell control optimization instruction; and S5, the physical entity PEM electrolytic bath executes the instruction. According to the method, the'electricity price-demand-equipment 'dynamic relation is planned as a whole from the global perspective, the process steps cover data-driven modeling, multi-target decision, digital twinning cooperation and closed-loop feedback, balance needs to be achieved among algorithm efficiency, model precision and engineering feasibility, and energy-economy-life multi-target cooperation is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of electrolytic hydrogen production, and specifically relates to an energy-economic collaborative optimization method for a PEM electrolyzer based on digital twins. Background Art

[0002] Driven by the global goal of carbon neutrality, hydrogen energy, as a zero-carbon energy carrier, has become one of the core paths for deep decarbonization of the energy system. The International Energy Agency (IEA) predicts that by 2050, the proportion of hydrogen energy in global energy consumption must increase from the current less than 1% to more than 13%, of which green hydrogen must account for more than 60%. Proton exchange membrane electrolyzers (PEMECs) are considered to be the core technical equipment for large-scale production of green hydrogen due to their high current density, fast start-stop response capabilities, and high adaptability to fluctuating renewable energy. However, current PEMEC technology still faces bottlenecks such as efficiency, lifespan, and economy. In particular, the problem of operation optimization under dynamic conditions is prominent, and breakthroughs are urgently needed through intelligent technology.

[0003] First, the intermittent nature of renewable energy causes dramatic fluctuations in input power. Traditional PEMEC control strategies, which primarily focus on steady-state operation, make it difficult to adjust parameters such as current density and temperature in real time, leading to efficiency losses and energy waste due to overvoltage and local overheating during dynamic processes. Frequent starts and stops and load fluctuations accelerate membrane electrode degradation. Secondly, the cost of green hydrogen is currently approximately $3 to $6 per kilogram, far exceeding that of gray hydrogen. Traditional control strategies are unable to effectively utilize differences in time-of-use electricity prices. Hydrogen demand forecasts are disconnected from production plans, leading to overproduction or supply shortages. Finally, in an integrated wind, solar, and hydrogen system, PEMEC needs to operate in coordination with equipment such as the power grid, hydrogen storage tanks, and fuel cells. However, existing technologies lack real-time interaction between the electrolyzer physical model and the power market and hydrogen demand models. They focus solely on single objectives of efficiency or economy and fail to achieve multi-objective coordination of energy, economy, and lifespan. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a PEM electrolyzer energy-economy collaborative optimization method based on digital twins. It coordinates the dynamic relationship of "electricity price-demand-equipment" from a global perspective. The process steps cover data-driven modeling, multi-objective decision-making, digital twin collaboration and closed-loop feedback, and achieves a balance between algorithm efficiency, model accuracy and engineering feasibility, realizing multi-objective collaboration of energy-economy-lifespan.

[0005] To achieve the above-mentioned multi-objective coordination of energy, economy, and lifespan, the present invention provides the following technical solution: a PEM electrolyzer energy-economy coordinated optimization method based on digital twin, comprising the following steps:

[0006] S1: Using the test data obtained from the standard test of a single PEM electrolyzer, combined with the existing membrane electrode mechanism mathematical model in the literature, the parameters of the membrane electrode mechanism mathematical model are calibrated. Through the dual correction method of test data and simulation data, a high-precision physical mechanism simulation model is established;

[0007] S2: Based on a high-precision physical mechanism simulation model, multi-physics field coupling simulation is used to build a high-precision digital twin system in virtual space that is highly consistent with the physical entity;

[0008] S3: Input electricity price fluctuation forecast data, hydrogen demand forecast data, and multi-dimensional data of equipment status, and achieve coordinated optimization analysis of the economic, safety, and sustainability of the PEM hydrogen production system through optimization decision-making algorithms;

[0009] S4: Based on the results of the collaborative optimization analysis of the economics, safety, and sustainability of the PEM hydrogen production system, the collaborative optimization analysis data is simulated through a high-precision digital twin system to generate PEM electrolyzer operation and control optimization instructions;

[0010] S5: The physical entity PEM electrolyzer executes the PEM electrolyzer operation control optimization instruction.

[0011] Furthermore, the standard tests of the PEM electrolyzer single cell in step S1 include: temperature test, pressure test, current density test, voltage test, hydrogen yield test, oxygen yield test, hydrogen purity analysis test, oxygen purity analysis test, and electrochemical impedance spectroscopy test.

[0012] Furthermore, the high-precision digital twin system in step S2 is a data-driven high-precision simulation model. Based on the digital twin framework, a high-precision digital twin system of the proton exchange membrane electrolyzer is built from five aspects: simulation physical layer, object data, dynamic model, functional module and application capability.

[0013] Furthermore, the electricity price fluctuation prediction data in step S3 includes: historical electricity market data, weather forecasts, grid load forecasts, and policy adjustment information.

[0014] Furthermore, the electricity price fluctuation prediction data uses a time series decomposition algorithm to separate the long-term trend and short-term fluctuation of electricity prices, and generates a probability distribution of electricity prices in the next 24 hours.

[0015] Furthermore, the hydrogen demand forecast data in step S3 includes: industrial user contracts, hydrogen refueling station orders, energy storage system scheduling plans, and hydrogen energy trading platform data.

[0016] Furthermore, the hydrogen demand forecast data is based on a federated learning framework that integrates multi-source data and constructs multiple demand scenarios to quantify the uncertainty of demand.

[0017] Furthermore, the multi-dimensional data of the equipment status in step S3 includes: electrolyzer health status data measured by the physical electrolyzer sensor, including membrane moisture content data and catalyst activity data, hydrogen storage tank remaining capacity data, and grid interaction power limit data.

[0018] Furthermore, the optimization decision algorithm in step S3 uses a weighted summation method to integrate multiple interrelated objectives into a comprehensive optimization problem, including the following steps:

[0019] Step 1: Construction of objective function

[0020] (1) Economic objective function

[0021] The economic objective function aims to minimize electricity costs. The electricity consumption of the electrolyzer is the main cost source of hydrogen production, and the operating power needs to be dynamically adjusted according to the time-of-use electricity price:

[0022]

[0023] C 电 (t): electricity price during period t ($ / kWh)

[0024] P(t): Power of the electrolyzer during period t (kW)

[0025] (2) Equipment life protection objective function

[0026] The equipment life protection objective function aims to minimize life loss. Frequent start-stop and load fluctuations will accelerate membrane electrode degradation, and their impact on life needs to be quantified:

[0027]

[0028] λ 寿命 (t): Life loss coefficient, related to current density fluctuation ΔI(t) and start-stop number N 启停 (t)Related.

[0029] (3) Hydrogen demand prediction objective function

[0030] The hydrogen demand forecast objective function aims to minimize the penalty for demand deviation, ensuring that hydrogen production meets downstream demand and avoiding supply shortages or surpluses:

[0031]

[0032] Q 需求 (t): Hydrogen demand during period t (kg)

[0033] Q 氢 (t): Actual hydrogen production (kg)

[0034] β: Shortage penalty coefficient ($ / kg).

[0035] (IV) Comprehensive objective function

[0036] Weighted integration of the above objectives:

[0037] min(ω1·Cost 电 +ω2·Cost 寿命 +ω3·Penalty 需求 )

[0038] ω (一)、 ω (二)、 ω3: Weight coefficient, reflecting the priority of different objectives, which needs to be determined through expert experience or sensitivity analysis.

[0039] Step 2: Setting hard constraints

[0040] (1) Hard constraints on supply and demand balance:

[0041] Q 氢 (t)+Q 储 (t-1)-Q 储 (t)≥Q 需求 (t)

[0042] (2) Hard constraints on equipment operation:

[0043] The minimum start-up and shutdown time of the electrolyzer is 30 minutes of cooling after shutdown before restarting;

[0044] Current density range: I min ≤I(t)≤I max .

[0045] (3) Hard constraints on hydrogen storage tank capacity:

[0046] 0≤Q 储 (t)≤Q 储max

[0047] (IV) Hard constraints for grid interaction:

[0048] Power ramp rate: |P(t)-P(t-1)|≤ΔP max ;

[0049] Maximum purchased power: P(t)≤P grid,max .

[0050] Step 3: Weighted Sum

[0051] By normalizing to the range of 0 to 1, standardize each target and assign weights ω according to priority (一)、 ω (二)、 ω3, converted into a single-objective optimization problem for solution.

[0052] Furthermore, the collaborative optimization analysis results in step S3 include: electricity price prediction results, demand probability distribution and equipment status matrix; the equipment status matrix includes: life attenuation coefficient.

[0053] Compared with the existing technology, the present invention provides a PEM electrolyzer energy-economy collaborative optimization method based on digital twin, which has the following beneficial effects:

[0054] First, it significantly reduces hydrogen production costs. By building a real-time interactive closed loop between physical equipment and virtual models, it can respond to electricity prices at all times, increase load rates during periods of low electricity prices (such as periods of excess wind power), and reduce the proportion of electricity costs.

[0055] Second, dynamically adjust production plans through hydrogen demand forecasts to reduce hydrogen storage equipment investment and hydrogen abandonment losses. By coupling electricity price signals, hydrogen demand forecasts, and equipment status, dynamic decision-making can be achieved from minute to monthly levels.

[0056] Third, wind, solar, and hydrogen will be used in synergy to smooth out fluctuations in renewable energy, improve the grid's absorption capacity, achieve hydrogen-electricity bidirectional regulation, switch to hydrogen storage and energy supply mode during peak hours of the grid, and participate in the power auxiliary service market;

[0057] Fourth, through digital twin optimization, the cost of green hydrogen is expected to be competitive with gray hydrogen, supporting the coordinated optimization of multi-electrolyzer clusters to adapt to 100-megawatt hydrogen production projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 Data flow diagram of energy-economy collaborative optimization method for PEM electrolyzer based on digital twin;

[0059] Figure 2 Figure 2 is a step-by-step diagram of the energy-economy co-optimization method for PEM electrolyzers based on digital twins. DETAILED DESCRIPTION

[0060] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0061] Please refer to the attached Figure 1 and attached Figure 2, a PEM electrolyzer energy-economy coordinated optimization method based on digital twin, comprising the following steps: S1: using test data obtained from a standard test on a single PEM electrolyzer cell, combined with a membrane electrode mechanism mathematical model already available in the literature, to calibrate the membrane electrode mechanism mathematical model parameters, and establish a high-precision physical mechanism simulation model through a dual correction method of test data and simulation data;

[0062] S2: Based on a high-precision physical mechanism simulation model, multi-physics field coupling simulation is used to build a high-precision digital twin system in virtual space that is highly consistent with the physical entity;

[0063] S3: Input electricity price fluctuation forecast data, hydrogen demand forecast data, and multi-dimensional data of equipment status, and achieve coordinated optimization analysis of the economic, safety, and sustainability of the PEM hydrogen production system through optimization decision-making algorithms;

[0064] S4: Based on the results of the collaborative optimization analysis of the economics, safety, and sustainability of the PEM hydrogen production system, the collaborative optimization analysis data is simulated through a high-precision digital twin system to generate PEM electrolyzer control optimization instructions;

[0065] S5: The physical entity PEM electrolyzer executes the PEM electrolyzer control optimization instruction to realize the closed loop of the physical entity decision optimization function.

[0066] As a preferred embodiment, the standard tests of the PEM electrolyzer cell in step S1 include: temperature test, pressure test, current density test, voltage test, hydrogen yield test, oxygen yield test, hydrogen purity analysis test, oxygen purity analysis test, and electrochemical impedance spectroscopy test. The electrochemical impedance spectroscopy test is used to analyze electrode process dynamics, double layer and diffusion, etc., and to study electrode materials, solid electrolytes, conductive polymers, and corrosion protection mechanisms.

[0067] As a preferred embodiment, the high-precision digital twin system in step S2 constructs a dynamic simulation model of the physical entity in a digital way with multiple dimensions, multiple time and space angles, multiple disciplines, and multiple physical quantities, so as to simulate and express the properties, behaviors, rules, etc. of the physical entity in the real environment. The digital twin system realizes the precise mapping of the physical entity in the digital space. It is a virtual model corresponding to the physical entity created by digital means, and realizes the full life cycle management of the physical entity based on real-time data.

[0068] The high-precision digital twin system is a data-driven high-precision simulation model. Based on the digital twin framework, a high-precision digital twin system of the proton exchange membrane electrolyzer is built from five aspects: simulation physical layer, object data, dynamic model, functional module and application capability.

[0069] 1. Simulate the physical layer

[0070] The physical entity layer is replaced by a high-precision physical mechanism simulation model as the simulated physical layer, forming a simulated physical layer that reflects the existing logical rules such as various operating logics, production processes, etc. within and between physical entities.

[0071] 2. Data Layer

[0072] The data in the data layer comes from the inherent data in the simulation physical layer and the multi-mode and multi-type operation data collected during the simulation operation.

[0073] 3. Model layer

[0074] The model layer is the digital twin system, which includes both digital models corresponding to known physical objects and a large number of data-driven models.

[0075] 4. Functional layer

[0076] The core element of the functional layer, "functional modules," refers to semi-autonomous subsystems formed by various models working together, either independently or in conjunction with each other. This semi-autonomy means that these functional modules can be independently designed and innovated, but they must adhere to common design rules to maintain a certain degree of consistency. This feature allows digital twin modules to be flexibly expanded, excluded, replaced, or modified, and can be recombined to achieve complex applications and form a mature and complete digital twin system.

[0077] 5. Capability layer

[0078] The digital twin system can provide external application capabilities to solve certain specific problems in specific application scenarios through the combination of functional modules. Because of the semi-autonomous characteristics of its internal models and modules, the formed patterns can be adaptively adjusted to a certain extent.

[0079] As a preferred embodiment, the electricity price fluctuation prediction data in step S3 includes: historical electricity market data, weather forecast, grid load forecast, and policy adjustment information;

[0080] Preferably, a time series decomposition algorithm is used to separate long-term trends and short-term fluctuations in electricity prices to generate a probability distribution of electricity prices for the next 24 hours; the time series decomposition algorithm refers to a basic model for constructing time series using the time series decomposition method. The time series decomposition method has been very useful for many years and reflects the development direction of economic phenomena over a long period of time.

[0081] Furthermore, the hydrogen demand forecast data in step S3 includes: industrial user contracts, hydrogen refueling station orders, energy storage system scheduling plans, and hydrogen energy trading platform data;

[0082] Preferably, multi-source data is integrated based on the federated learning framework, and multiple demand scenarios are constructed to quantify the uncertainty of demand.

[0083] Furthermore, the multi-dimensional data of the equipment status in step S3 includes: electrolyzer health status data measured by the physical electrolyzer sensor, including membrane moisture content data and catalyst activity data, hydrogen storage tank remaining capacity data, and grid interaction power limit data.

[0084] As a preferred embodiment, the optimization decision algorithm in step S3 is to integrate multiple interrelated objectives into a comprehensive optimization problem using a weighted summation method, including the following steps:

[0085] Step 1: Construction of objective function

[0086] (1) Economic objective function

[0087] The economic objective function aims to minimize electricity costs. The electricity consumption of the electrolyzer is the main cost source of hydrogen production, and the operating power needs to be dynamically adjusted according to the time-of-use electricity price:

[0088]

[0089] C 电 (t): electricity price during period t ($ / kWh)

[0090] P(t): Power of the electrolyzer during period t (kW)

[0091] (2) Equipment life protection objective function

[0092] The equipment life protection objective function is to minimize the life loss, frequent start and stop and load fluctuation

[0093] Movement will accelerate membrane electrode degradation, and its impact on life needs to be quantified:

[0094]

[0095] λ 寿命 (t): Life loss coefficient, related to current density fluctuation ΔI(t) and start-stop number N 启停 (t)Related.

[0096] (3) Hydrogen demand prediction objective function

[0097] The hydrogen demand forecast objective function aims to minimize the penalty for demand deviation, ensuring that hydrogen production meets downstream demand and avoiding supply shortages or surpluses:

[0098]

[0099] Q 需求 (t): Hydrogen demand during period t (kg)

[0100] Q 氢 (t): Actual hydrogen production (kg)

[0101] β: Shortage penalty coefficient ($ / kg).

[0102] (IV) Comprehensive objective function

[0103] Weighted integration of the above objectives:

[0104] min(ω1·Cost 电 +ω2·Cost 寿命 +ω3·Penalty 需求 )

[0105] ω (一)、 ω (二)、 ω3: is the weight coefficient, reflecting the priority of different goals, which needs to be determined through expert experience

[0106] or determined by sensitivity analysis.

[0107] Step 2: Setting hard constraints

[0108] (1) Hard constraints on supply and demand balance:

[0109] Q 氢 (t)+Q 储 (t-1)-Q 储 (t)≥Q 需求 (t)

[0110] Q 氢 (t): The amount of hydrogen produced at time t

[0111] Q 储( t): Hydrogen storage at time t

[0112] Q 储 (t-1): Hydrogen reserve at time t-1

[0113] Q 需求 (t): hydrogen demand at time t;

[0114] (2) Hard constraints on equipment operation:

[0115] The minimum start-up and shutdown time of the electrolyzer is 30 minutes of cooling after shutdown before restarting;

[0116] Current density range: I min ≤I(t)≤I max .

[0117] (3) Hard constraints on hydrogen storage tank capacity:

[0118] 0≤Q 储 (t)≤Q 储 max

[0119] (IV) Hard constraints for grid interaction:

[0120] Power ramp rate: |P(t)-P(t-1)|≤ΔP max ;

[0121] Maximum purchased power: P(t)≤P grid,max ;

[0122] P grid,max The maximum power allowed by the power grid for purchasing electricity.

[0123] Step 3: Weighted Sum

[0124] By normalizing to the range of 0 to 1, standardize each target and assign weights ω according to priority (一)、 ω (二)、 ω3, converted into a single-objective optimization problem for solution.

[0125] As a preferred embodiment, the collaborative optimization analysis results in step S3 include:

[0126] Electricity price forecast results, demand probability distribution, and equipment status matrix;

[0127] Furthermore, the device state matrix includes: a life attenuation coefficient.

[0128] As a preferred embodiment, the high-precision digital twin system analysis in step S4 includes: digital twin system calibration, feeding back actual operation data to the digital twin system, and the digital twin system updating model parameters;

[0129] As a preferred embodiment, in step S4, the PEM electrolyzer control optimization instruction is transmitted to the real-time controller via the industrial communication protocol, and the physical entity PEM electrolyzer executes the PEM electrolyzer control optimization instruction.

[0130] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0131] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A PEM electrolyzer energy-economy collaborative optimization method based on digital twin, characterized by: The following steps are involved: S1: Using the test data obtained from the standard test of a single PEM electrolyzer cell, combined with the existing membrane electrode mechanism mathematical model in the literature, the parameters of the membrane electrode mechanism mathematical model are calibrated. Through the dual correction method of test data and simulation data, a high-precision physical mechanism simulation model is established; S2: Based on a high-precision physical mechanism simulation model, multi-physics field coupling simulation is used to build a high-precision digital twin system in virtual space that is highly consistent with the physical entity; S3: Input electricity price fluctuation forecast data, hydrogen demand forecast data, and multi-dimensional data of equipment status, and achieve coordinated optimization analysis of the economic, safety, and sustainability of the PEM hydrogen production system through optimization decision-making algorithms; S4: Based on the collaborative optimization analysis results of the economic, safety, and sustainability of the PEM hydrogen production system, the collaborative optimization analysis results are simulated and run through a high-precision digital twin system to generate PEM electrolyzer operation and control optimization instructions; S5: The physical entity PEM electrolyzer executes the PEM electrolyzer operation control optimization instruction.

2. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 1, characterized in that: The standard tests of the PEM electrolyzer single cell in step S1 include: temperature test, pressure test, current density test, voltage test, hydrogen yield test, oxygen yield test, hydrogen purity analysis test, oxygen purity analysis test, and electrochemical impedance spectroscopy test.

3. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 1, characterized in that: The high-precision digital twin system in step S2 is a data-driven high-precision simulation model, which is a high-precision digital twin system of a proton exchange membrane electrolyzer built from five aspects: simulation physical layer, object data, dynamic model, functional module and application capability.

4. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 1, characterized in that: The electricity price fluctuation prediction data in step S3 includes: historical electricity market data, weather forecast, grid load forecast, and policy adjustment information.

5. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 4 is characterized in that: The electricity price fluctuation prediction data uses a time series decomposition algorithm to separate the long-term trend and short-term fluctuation of electricity prices, and generates a probability distribution of electricity prices in the next 24 hours.

6. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 1, characterized in that: The hydrogen demand forecast data in step S3 includes: industrial user contracts, hydrogen refueling station orders, energy storage system scheduling plans, and hydrogen energy trading platform data.

7. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 6, characterized in that: The hydrogen demand forecast data is based on a federated learning framework that integrates multi-source data and constructs multiple demand scenarios to quantify the uncertainty of demand.

8. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 1, characterized in that: The multi-dimensional data of the equipment status in step S3 includes: electrolyzer health status data measured by the physical electrolyzer sensor, including membrane moisture content data and catalyst activity data, hydrogen storage tank remaining capacity data, and grid interaction power limit data.

9. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 1, characterized in that: The optimization decision algorithm in step S3 uses a weighted summation method to integrate multiple interrelated objectives into a comprehensive optimization problem, including the following steps: Step 1: Construction of objective function (1) The economic objective function needs to dynamically adjust the operating power according to the power consumption of the electrolyzer and the time-of-use electricity price: (2) The equipment life protection objective function aims to minimize life loss. Frequent start-stop and load fluctuations will accelerate membrane electrode degradation, and their impact on life needs to be quantified: λ 寿命 (t): Life loss coefficient, related to current density fluctuation ΔI(t) and start-stop number N 启停 (t) relevant; (3) Hydrogen demand prediction objective function The hydrogen demand forecast objective function aims to minimize the demand deviation penalty, ensure that hydrogen production meets downstream demand, and avoid supply shortages or surpluses: Q 需求 (t): Hydrogen demand during period t Q 氢 (t): Actual hydrogen production β: shortage penalty coefficient; (IV) Comprehensive objective function The above economic objective function, equipment life protection objective function and hydrogen demand prediction objective function are combined Weighted integration: min(ω1·Cost 电 +ω2·Cost 寿命 +ω3·Penalty 需求 ) ω (一)、 ω (二)、 ω3: is the weight coefficient, reflecting the priority of different goals, which needs to be determined through expert experience or determined by sensitivity analysis; Step 2: Setting hard constraints (1) Hard constraints on supply and demand balance: Q 氢 (t)+Q 储 (t-1)-Q 储 (t)≥Q 需求 (t); (2) Hard constraints on equipment operation: The minimum start-up and shutdown time of the electrolyzer is 30 minutes of cooling after shutdown before restarting; Current density range: I min ≤I(t)≤I max ; (3) Hard constraints on hydrogen storage tank capacity: 0≤Q 储 (t)≤Q 储max (IV) Hard constraints for grid interaction: Power ramp rate: |P(t)-P(t-1)|≤ΔP max ; Maximum purchased power: P(t)≤P grid,max ; Step 3: Weighted Sum By normalizing to the range of 0 to 1, standardize each target and assign weights ω according to priority (一)、 ω (二)、 ω3, converted into a single-objective optimization problem for solution.

10. The energy-economy collaborative optimization method for PEM electrolyzers based on digital twins according to claim 1, characterized in that: The collaborative optimization analysis results in step S3 include: electricity price forecast results, demand probability distribution and equipment status matrix; the equipment status matrix includes: life attenuation coefficient.

Images