Electrolytic cell dynamic modeling method and system, storage medium, computer device

Abstract

The application discloses an electrolytic cell dynamic modeling method and system, a storage medium and a computer device. The method comprises the following steps: constructing a full-scale dynamic model of an electrolytic cell, including a flow channel gas-liquid two-phase flow control equation, establishing a main electrode current-heat coupling, a diaphragm mass transfer-heat transfer control equation, a multi-field coupling control equation of a porous electrode, embedding a Nernst equation and a Butler-Volmer equation to associate reaction kinetics parameters, and introducing a conductivity micro-correction model. Reaction kinetics parameters and flow mass transfer parameters are corrected through a heuristic high-dimensional space variable sensitivity reduction gradient regression algorithm. Based on real-time operation data, a multi-scale model is coupled to solve the current operation state of the electrolytic cell and predict future trends. By focusing on the multi-field coupling characteristics of the electrolytic cell, a multi-scale high-fidelity model is constructed, which can improve the speed of predicting the state parameters of the process equipment (electrolytic cell) and effectively support the efficient operation of the hydrogen production process system.

Description

Technical Field

[0001] This application relates to the field of electrolytic hydrogen production technology, and in particular to a dynamic modeling method and system for electrolyzers, a storage medium, and a computer device. Background Technology

[0002] Against the backdrop of global carbon neutrality goals and energy restructuring, the current situation where fossil fuels account for over 80% of energy consumption continues to exacerbate climate problems. Meanwhile, renewable energy sources such as wind and solar power are difficult to support a stable energy supply due to their intermittency and volatility. Therefore, hydrogen energy, with its advantages of high energy density, zero-carbon combustion, and adaptability to multiple scenarios, has become a new frontier in global energy competition.

[0003] Although my country has a world-leading installed capacity of wind and solar power and a natural advantage in green electricity hydrogen production, the volatility of new energy sources leads to unstable loads and frequent start-ups and shutdowns of electrolyzers, which exacerbates equipment wear and tear and operation and maintenance costs. Summary of the Invention

[0004] In view of this, this application provides a dynamic modeling method and system for electrolyzers, a storage medium, and a computer device. It utilizes multi-scale distributed parametric modeling technology for hydrogen production equipment, supplemented by a heuristic high-dimensional variable sensitivity reduction gradient regression algorithm (HHDSDRGRA) and automatic calibration technology for process operation data. Focusing on the multi-field coupling characteristics of the electrolyzer's "mechanical-electrical-thermal-chemical" systems, it constructs a multi-scale high-fidelity model (electrolyzer dynamic model control equations). The model's expressive power and its ability to fit actual equipment operating parameters are optimized, thereby generating a high-precision, long-cycle, real-time hydrogen production process equipment operating parameter prediction model at a lower cost. This model can quickly predict process equipment state parameters, effectively supporting the efficient operation of the hydrogen production process system.

[0005] According to one aspect of this application, a method for dynamic modeling of an electrolytic cell is provided, the method comprising: The mass conservation equation, energy conservation equation and momentum conservation equation of the fluid in the integrated electrolyzer are expanded along the flow direction of the fluid in the electrolyzer to construct the flow channel model control equation. The flow channel model control equation characterizes the gas-liquid two-phase flow phenomenon that occurs in the electrolyzer in the macroscopic dimension. A current conservation equation is constructed to quantify the voltage loss of the main electrode plate in the electrolytic cell during current transmission, and an energy conservation equation is constructed to characterize the temperature change trend of the main electrode plate over time. The equations are expanded along the thickness direction of the anode and cathode in the main electrode plate, and the current conservation equation and the energy conservation equation are coupled to obtain the main electrode plate model control equation. The main electrode plate model control equation characterizes the main electrode plate in the mesoscopic dimension. In the porous electrode of the integrated electrolytic cell, the mass conservation equation, energy conservation equation, momentum conservation equation, and electron conservation equation of each substance are used to construct the governing equation of the porous electrode model. The governing equation of the porous electrode model characterizes the porous electrode at the microscopic level. When constructing the governing equation of the porous electrode model, the Nernst equation and the Butler-Volmer equation are introduced to characterize the electrochemical reaction process in the porous electrode, and the reaction kinetic parameters in the electrochemical reaction process are correlated with various conservation equations. Based on the energy conservation equation, electron conservation equation, and ion conservation equation of the membrane in the electrolyzer, as well as the transmembrane transport mechanism of various substances in the membrane, the membrane model governing equation is constructed. The membrane model governing equation characterizes the material transport and energy transfer process in the membrane in the mesoscopic dimension. Considering the synergistic effects of ambient temperature, ion concentration, and solution temperature on electronic conductivity during electrolytic cell operation, a microscopic correction model governing equation for conductivity is constructed. A heuristic high-dimensional space variable sensitivity reduced-order gradient regression algorithm is introduced to correct reaction kinetic parameters and flow mass transfer parameters in real time. The control equations of the flow channel model, the main electrode model, the porous electrode model, the diaphragm model, and the conductivity correction model are coupled in a full-scale manner to construct the control equations of the electrolyzer dynamic model. Based on the real-time operating parameters of the electrolyzer, the current operating state and future operating state of the electrolyzer can be obtained by solving the control equations of the electrolyzer dynamic model. The operating state is characterized by the parameter values ​​obtained by solving the control equations of the electrolyzer dynamic model.

[0006] According to another aspect of this application, a dynamic modeling system for an electrolytic cell is provided, the system comprising: The model building module is used to integrate the mass conservation equation, energy conservation equation and momentum conservation equation of the fluid in the electrolyzer, expand along the flow direction of the fluid in the electrolyzer, and construct the flow channel model control equation. The flow channel model control equation characterizes the gas-liquid two-phase flow phenomenon that occurs in the electrolyzer in the macroscopic dimension. The model building module is also used to construct the current conservation equation for quantifying the voltage loss of the main electrode plate in the electrolytic cell during the current transmission process, and to construct the energy conservation equation characterizing the temperature change trend of the main electrode plate over time. The equation is expanded along the thickness direction of the anode and cathode in the main electrode plate, and the current conservation equation and the energy conservation equation are coupled to obtain the main electrode plate model control equation. The main electrode plate model control equation characterizes the main electrode plate in the mesoscopic dimension. The model building module is also used to synthesize the mass conservation equation, energy conservation equation, momentum conservation equation, and electron conservation equation of each substance in the porous electrode of the electrolyzer, and to construct the control equation of the porous electrode model. The control equation of the porous electrode model characterizes the porous electrode at the microscopic level. When constructing the control equation of the porous electrode model, the Nernst equation and the Butler-Volmer equation are introduced to characterize the electrochemical reaction process in the porous electrode, and the reaction kinetic parameters in the electrochemical reaction process are correlated with various conservation equations. The model building module is also used to construct the membrane model control equations based on the energy conservation equation, electron conservation equation and ion conservation equation of the membrane in the electrolyzer, as well as the transmembrane transport mechanism of each substance in the membrane, along the membrane thickness direction. The membrane model control equations characterize the material transport and energy transfer process in the membrane in the mesoscopic dimension. The model building module is also used to consider the synergistic effects of ambient temperature, ion concentration and solution temperature on electronic conductivity during the operation of the electrolyzer, and to build the control equations for the conductivity micro-correction model. The multiphysics model fusion module is used to introduce a heuristic high-dimensional spatial variable sensitivity reduced-order gradient regression algorithm to correct reaction kinetic parameters and flow mass transfer parameters in real time. It couples the control equations of the flow channel model, the main electrode model, the porous electrode model, the diaphragm model, and the conductivity correction model at all scales to construct the control equations of the electrolyzer dynamic model. Based on the real-time operating parameters of the electrolyzer, the current operating state and future operating state of the electrolyzer can be obtained by solving the control equations of the electrolyzer dynamic model. The operating state is represented by the parameter values ​​obtained by solving the control equations of the electrolyzer dynamic model.

[0007] According to another aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the above-described dynamic modeling method for electrolytic cells.

[0008] According to another aspect of this application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the above-described dynamic modeling method for electrolytic cells.

[0009] By employing the above technical solutions, this application provides a dynamic modeling method and system for electrolyzers, a storage medium, and computer equipment. First, a distributed parameter method is used to divide the model into micro-segments, establishing energy conservation, mass conservation, momentum conservation, and electrochemical reaction mechanism models at multiple physical scales. These models include flow channel models, main electrode models, porous electrode models, and membrane models. Sub-models for gas-liquid separation and transmembrane transport are integrated, and the HHDSDRGRA algorithm is introduced to correct reaction kinetic parameters and flow mass transfer parameters. Automated calibration is performed using measured process data to reduce model errors under different loads and temperatures. Through the aforementioned data processing, the multi-field coupling characteristics of the electrolyzer's "mechanical-electrical-thermal-chemical" processes are focused to construct a multi-scale high-fidelity model. The proposed multi-scale distributed parameterized modeling technology for hydrogen production equipment, supplemented by the HHDSDRGRA algorithm and automatic calibration technology for process operation data, simultaneously solves the problems of modeling accuracy, real-time prediction, and modeling cost for hydrogen production equipment, achieving a comprehensive improvement in modeling accuracy, modeling cost, and long-term prediction data stability.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart illustrating a dynamic modeling method for an electrolytic cell provided in an embodiment of this application is shown. Figure 2 A schematic diagram of the structure of a dynamic modeling system for an electrolytic cell provided in an embodiment of this application is shown. Detailed Implementation

[0012] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0013] This embodiment provides a method for dynamic modeling of electrolytic cells, such as... Figure 1 As shown, the method includes: Step 101: Combine the mass conservation equation, energy conservation equation and momentum conservation equation of the fluid in the electrolyzer, expand them along the flow direction of the fluid in the electrolyzer, and construct the flow channel model control equation. The flow channel model control equation characterizes the gas-liquid two-phase flow phenomenon that occurs in the electrolyzer in the macroscopic dimension. Step 102: Construct a current conservation equation for quantifying the voltage loss of the main electrode plate in the electrolytic cell during current transmission, and construct an energy conservation equation to characterize the temperature change trend of the main electrode plate over time. Expand along the thickness direction of the anode and cathode in the main electrode plate, and couple the current conservation equation and the energy conservation equation to obtain the main electrode plate model control equation. The main electrode plate model control equation characterizes the main electrode plate in the mesoscopic dimension. Step 103: Based on the mass conservation equation, energy conservation equation, momentum conservation equation, and electron conservation equation of each substance in the porous electrode of the electrolytic cell, construct the control equation of the porous electrode model. The control equation of the porous electrode model characterizes the porous electrode at the microscopic level. When constructing the control equation of the porous electrode model, the Nernst equation and the Butler-Volmer equation are introduced to characterize the electrochemical reaction process in the porous electrode, and the reaction kinetic parameters in the electrochemical reaction process are correlated with various conservation equations. Step 104: Based on the energy conservation equation, electron conservation equation and ion conservation equation of the membrane in the electrolytic cell, as well as the transmembrane transport mechanism of each substance in the membrane, expand along the membrane thickness direction to construct the membrane model control equation. The membrane model control equation characterizes the material transport and energy transfer process in the membrane in the mesoscopic dimension. Step 105: Considering the synergistic effects of ambient temperature, ion concentration, and solution temperature on electronic conductivity during the operation of the electrolyzer, construct the control equations for the conductivity micro-correction model. Step 106 introduces a heuristic high-dimensional space variable sensitivity reduced-order gradient regression algorithm to correct the reaction kinetic parameters in real time. The control equations of the flow channel model, the main electrode model, the porous electrode model, the diaphragm model, and the conductivity correction model are coupled in a full-scale manner to construct the control equations of the electrolyzer dynamic model. Based on the real-time operating parameters of the electrolyzer, the current operating state and future operating state of the electrolyzer can be obtained by solving the control equations of the electrolyzer dynamic model. The operating state is characterized by the parameter values ​​obtained by solving the control equations of the electrolyzer dynamic model.

[0014] Currently, the traditional experience-driven model can no longer meet the needs of large-scale hydrogen energy development, and there is an urgent need to overcome bottlenecks in efficiency, safety, and economy through advanced technologies. Developing high-fidelity simulation models for electrolyzers can accurately capture dynamic changes in hydrogen production parameters, reproduce complex operating conditions, and provide short-, medium-, and long-term multi-timescale predictive data, offering technical support for equipment research and development, design, operation and maintenance, and scheduling.

[0015] Traditional macroscopic modeling methods based on lumped parameter methods neglect too many structural details of equipment, failing to accurately capture the actual operating state and generally resulting in low accuracy. While full 3D equipment simulation based on distributed modeling technology can provide detailed parameters of key internal equipment, it involves large computational loads, poor real-time performance, and is unsuitable for large-scale applications. Statistical models based solely on process operation data can achieve high data real-time performance, but require long-term operational data support, leading to long modeling cycles. Furthermore, due to the scarcity of fault data, they cannot effectively cover the equipment's operating state and are prone to outputting data that does not conform to physical reality. Further improvements are needed in modeling cost, modeling cycle, model accuracy, and the stability of long-term predictive data.

[0016] The above embodiments of this application may include the following models: 1. Flow channel model (governing equations for the flow channel model). This is a macroscopic model. The flow channel model is a gas-liquid two-phase flow model, considering the mass conservation equation, energy conservation equation, and momentum conservation equation for the fluid. The related models are parametric models, expanded along the flow direction x.

[0017] 2. Main Plate Model (Governing Equations for the Main Plate Model). This model is a mesoscopic model. Current and heat transfer occur within the main plate, involving energy conservation and current conservation models. The energy conservation model describes the temperature change of the main plate over time, while the current conservation model describes the voltage loss caused by the current flow. The relevant models are parametric, expanded along the z-axis of the anode and cathode thickness.

[0018] 3. Porous Electrode Model (Governing Equations for Porous Electrode Model). This model is a microscopic model. It considers the mass conservation equations, energy conservation equations, momentum conservation equations, and electron conservation equations for each substance in the porous electrode. The electrochemical reaction processes involving each substance are described using the Nernst equation and the Butler-Volmer equation, and correlated with the relevant conservation equations.

[0019] 4. Diaphragm Model (Governing Equations for the Diaphragm Model). This model is a mesoscopic model. It considers the energy conservation equations, electron conservation equations, and ion conservation equations within the diaphragm, as well as the transmembrane transport models for various substances. This is a parametric model, expanded along the membrane thickness direction z.

[0020] 5. Conductivity Correction Model (Control Equations of Microscopic Conductivity Correction Model). This model is a microscopic model. It considers the effect of temperature to perform a refined modeling of electronic conductivity. It also considers the influence of factors such as ion concentration and solution temperature to perform a refined modeling of ionic conductivity.

[0021] Furthermore, the full-scale coupled electrolyzer dynamic model of this application can accurately reproduce the operational essence of the electrolyzer from multiple dimensions. The macroscopic flow channel model can accurately capture the evolution of the gas-liquid two-phase flow, providing reliable flow field support for overall operation, effectively avoiding local efficiency losses caused by uneven flow field distribution, and ensuring the stability and energy efficiency of the fluid transport process. The mesoscopic master plate model focuses on voltage loss and temperature dynamic response during current transmission, accurately characterizing the energy conversion and thermal characteristics of the plate, and timely detecting heat accumulation or voltage anomalies, providing quantitative basis for plate thermal management and energy efficiency optimization. The microscopic porous electrode model deeply binds the electrochemical reaction mechanism and multi-field processes, meticulously reproducing the reaction and mass transfer laws inside the electrode, accurately controlling the matching degree of reaction rate and mass transfer, providing precise guidance for the optimized design of electrode structure, and helping to improve reaction conversion rate and selectivity. The mesoscopic membrane model clearly presents the mass migration and energy transfer processes within the membrane, enabling timely detection of performance degradation caused by poor mass transfer and ensuring the balance of matter and energy within the electrolyzer. The conductivity micro-correction model comprehensively considers the synergistic effects of multiple operating conditions on conductivity, making parameter characterization more closely aligned with actual operating scenarios and significantly improving model calculation accuracy. Combined with the HHDSDRGRA algorithm for real-time correction of reaction kinetic parameters, the fully coupled dynamic model enables precise perception of the current operating status of the electrolyzer and early prediction of future trends, providing reliable support for real-time optimization of operating parameters and early avoidance of potential faults, comprehensively improving the electrolyzer's operating efficiency, stability, and service life.

[0022] Specifically, when coupling the control equations of the flow channel model, the main electrode model, the porous electrode model, the diaphragm model, and the conductivity correction model across the entire scale, the fusion method is as follows: 1. Based on the iterative method, different calculation step sizes are used to solve the simultaneous equations of all models.

[0023] Specifically, to achieve full-scale coupling of the five types of models, we can first define the parameter transfer logic between models of different scales and establish cross-scale interaction links: Using the solution temperature, ion concentration, and ambient temperature output from the flow channel model as inputs, the corrected electronic conductivity is calculated in real time. ionic conductivity The parameters are synchronously transmitted to the main electrode model (electron conservation equation), the porous electrode model (electron / ion conservation equation), and the membrane model (electron / ion conservation equation), providing accurate conductivity parameters for all models involving electrical transport.

[0024] The voltage drop output by the main electrode model serves as the electrical input boundary of the porous electrode model; the heat flux generated by the porous electrode model serves as the thermal input boundary of the main electrode model; the ion flux received by the membrane model from the porous electrodes on both sides serves as the input, and the output transmembrane ion potential and flux serve as the boundary conditions of the porous electrode.

[0025] The fluid temperature and component concentration output by the flow channel model serve as the boundary of the reaction medium in the porous electrode model; the gas-liquid products and heat flux generated by the porous electrode model serve as the input to the flow channel model, updating the gas-liquid two-phase flow state within the flow channel.

[0026] Next, since the spatiotemporal scales of different models differ significantly (the flow channel is macroscopic, the porous electrode is microscopic, and the main electrode / diaphragm is mesoscopic), a multi-step iterative simultaneous solution method can be adopted. Specifically: Initialization and parameter assignment: Input the real-time operating parameters of the electrolyzer, such as ambient temperature, initial current density, and initial electrolyte concentration, to assign initial parameters to all models; at the same time, initialize the reaction kinetic parameters of the HHDSDRGRA algorithm to prepare for subsequent real-time correction.

[0027] Multi-scale pre-computation sets differentiated computational step sizes based on the scale characteristics of the model, achieving efficient multi-scale solutions. For example: Microscopic small-step calculation: Solve the porous electrode model and conductivity correction model simultaneously with the smallest time step in milliseconds: First, use the conductivity correction model to correct the conductivity according to the current temperature and ion concentration, and substitute it into the electron / ion conservation equation of the porous electrode; at the same time, call the HHDSDRGRA algorithm to correct the reaction kinetic parameters according to the real-time running parameters, and solve for the microscopic parameters of the porous electrode such as reaction rate, ion / electron flux, and heat generation.

[0028] Mesoscopic time step calculation: Solve the main electrode model + membrane model with a medium time step of seconds: The main electrode model receives the heat flux and corrected conductivity of the porous electrode to calculate the voltage drop and temperature distribution; the membrane model receives the ion flux and corrected conductivity of the porous electrodes on both sides to calculate the transmembrane ion transport and heat transfer process, and obtain the mesoscopic material and energy parameters.

[0029] Macroscopic large step size calculation: Solve the flow channel model with a maximum time step of minutes: Take the gas-liquid products output by the porous electrode and the heat flux of the main electrode plate as input, and combine the mass / energy / momentum conservation equations of gas-liquid two-phase flow to solve the macroscopic parameters such as fluid temperature, concentration, and pressure distribution in the flow channel.

[0030] Next, cross-scale iterative correction and convergence judgment backpropagate the output parameters of each scale model to the associated model for iterative correction: The temperature and concentration parameters of the flow channel are fed back to the porous electrode model and the conductivity correction model to update the medium boundary for the next round of microscopic calculations; The voltage drop of the main electrode plate and the ionic potential of the membrane are fed back to the porous electrode model to update the boundary conditions of the electrochemical reaction. The conductivity correction model, based on the new temperature and concentration parameters, re-corrects the conductivity and synchronizes it to all associated models. This "segmented calculation - cross-scale feedback" cycle is repeated until the error of all parameters between two adjacent iterations is less than a set threshold (e.g., 10). -6 The model is determined to be converged, and the state parameters of the electrolytic cell under the current operating conditions are obtained.

[0031] After each round of convergence calculation, the HHDSDRGRA algorithm is used to correct the reaction kinetic parameters again, based on the new real-time operating parameters of the electrolyzer, such as real-time current and ambient temperature fluctuations. This updates the input boundaries of all models and enters the next round of full-scale coupled solution, thus realizing the real-time update of the dynamic model.

[0032] To address the full-scale linkage effect of the HHDSDRGRA algorithm, during the iteration process, the HHDSDRGRA algorithm can correct the reaction kinetic parameters in real time based on the deviation between the calculated values ​​of the porous electrode model and the actual operating data. The change of these parameters will be transmitted to the diaphragm, main electrode plate, and flow channel models through the ion / electron flux of the porous electrode, realizing the dynamic linkage of parameters of the full-scale model and ensuring the real-time performance and accuracy of the model when the electrolyzer operating conditions change.

[0033] Optionally, in step 101, the flow mass transfer parameters include a first flow resistance coefficient and a second flow resistance coefficient, and the mass conservation equation in the flow channel model governing equations is: ， The energy conservation equation is: , The momentum conservation equation is: , Along the flow direction of the fluid in the electrolytic cell, This represents the volume fraction of the gas / liquid phase. The density of the gas / liquid phase. The mass fraction of component i in the gas / liquid phase. For the mass transfer source term of component i in the gas / liquid phase, For gas / liquid phase velocity, The enthalpy of the gas / liquid phase mass ratio. The heat flux density between the gas / liquid phase and the main electrode plate within the flow channel. The heat flux density between the gas / liquid phase and the porous electrode plate within the flow channel. For fluid pressure, For fluid velocity, For fluid dynamic viscosity, These are the first flow resistance coefficient and the second flow resistance system, respectively. is the fluid density, t is time, and p is the phase, used to distinguish between the gas and liquid phases in gas-liquid two-phase flow phenomena.

[0034] In the above embodiments of this application, ,unit , ,unit , ,unit , ,unit , ,unit , ,unit , ,unit , ,unit , ,unit , ,unit , ,unit , ,unit .

[0035] Specifically, the process of constructing the control equations for the flow channel model involves, for example, constructing three types of equations along the fluid flow direction of the electrolyzer, taking into account the characteristics of the gas-liquid two-phase flow: the mass conservation equation incorporates the spatiotemporal changes in the volume fraction, density, and component mass fraction of the gas and liquid phases, superimposed with mass transfer source terms, reflecting the dynamic equilibrium of the two-phase components; the energy conservation equation considers the spatiotemporal evolution of the enthalpy of the gas and liquid phases, coupled with heat exchange terms with the main electrode and porous electrode, reflecting the heat transfer within the flow channel; and the momentum conservation equation relates to the dynamic changes in fluid pressure and velocity, introducing flow resistance coefficients and fluid properties to characterize the flow resistance and dynamic characteristics of the two-phase flow. These three equations together constitute the macroscopic control equations for the gas-liquid two-phase flow.

[0036] When solving the dynamic model of an electrolyzer in a full-scale coupled manner, the flow channel model can output state parameters such as flow field pressure, temperature, and two-phase distribution in real time, providing accurate flow field boundary conditions for other models such as porous electrodes and main plates. It supports the quantitative characterization of the flow field, thermal state, and other dimensions of the current operating state of the electrolyzer, and can also infer the future flow field evolution trend based on real-time parameters, assisting in the prediction of the subsequent operating state of the electrolyzer, and providing the core flow field foundation for the dynamic solution of the overall model.

[0037] Optionally, in step 102, the current conservation equation in the governing equations of the main plate model is: , The energy conservation equation is: , The thickness is along the anode and cathode direction, and T is the ambient temperature. For specific heat capacity, For density, For electron current density, Electron conductivity, The thermal conductivity of the main electrode material Let t be the electron potential, and t be time.

[0038] In the above embodiments of this application, the current conservation equation is also the current and potential equation, T, unit , ,unit , ,unit , ,unit , ,unit , ,unit , ,unit .

[0039] Specifically, regarding the construction process of the governing equations for the main electrode model, the derivation proceeds along the thickness direction of the anode and cathode of the main electrode. First, a current conservation equation is constructed for the voltage loss during current transmission: based on the steady-state transmission characteristics of electron current within the main electrode, the partial derivative of the electron current density along the thickness direction is assumed to be zero. Combining the Ohmic relationship of electron conductivity, the electron current density is defined through the correlation between electron conductivity and potential gradient, thereby quantifying the voltage loss distribution during current transmission within the electrode. Changes in the potential gradient directly correspond to the degree of voltage loss at different locations. Subsequently, an energy conservation equation for temperature dynamics is constructed: considering the heat capacity and density parameters of the main electrode, the temperature change over time is characterized, while incorporating the heat conduction term along the thickness direction, and combining it with the Joule heat generation term generated during current transmission, to depict the real-time evolution trend of the electrode temperature. Then, the two equations are coupled, correlating the electron current density obtained from the current conservation equation to the Joule heat term in the energy conservation equation, forming the governing equations for the main electrode model that accurately characterize the voltage loss and temperature dynamics of the main electrode in the mesoscopic dimension.

[0040] When simultaneously solving the dynamic model of the electrolyzer, the main electrode model can accurately output the voltage loss distribution and real-time temperature status of the main electrode, providing key mesoscopic boundary conditions for models at other scales: it transmits the potential and temperature boundaries on the electrode side to the porous electrode model, and provides the basic parameters for heat exchange between the electrode and the flow channel model. Simultaneously, based on the real-time operating parameters of the electrolyzer, it can intuitively reflect the current energy efficiency loss and thermal safety status of the main electrode, and can also predict future trends in temperature and voltage loss, assisting in judging the overall energy efficiency level and thermal stability of the electrolyzer. This provides core evidence for predicting whether the electrolyzer will experience thermal overload, voltage anomalies, or other faults, supporting the accurate characterization of the current and future operating states of the electrolyzer, and making the solution results of the full-scale model more consistent with the actual mesoscopic characteristics of operation.

[0041] Optionally, in step 103, the flow mass transfer parameters include the permeability of the porous medium, and the electron conservation equations in the porous electrode model include the electron current and potential equation, the ion current and potential equation, and the overpotential calculation equation. The mass conservation equation is: , The energy conservation equation is: , The momentum conservation equation is: , The equations for electron current and potential are: , , The equations for ion current and potential are as follows: , , The equation for calculating overpotential is: ， , The Nernst equation is: , The Butler-Volmer equation is: , The porosity of the porous electrode. This represents the volume fraction of the gas / liquid phase. The density of the gas / liquid phase. The mass fraction of component i in the gas / liquid phase. Let i be the molecular diffusion flux of component i. For the gas / liquid phase fluid velocity, In the direction of thickness of the anode and cathode, Let i be the stoichiometric coefficient of component i in the electrochemical reaction. The rate of electron generation in an electrochemical reaction. Where is the number of electrons transferred in the reaction, and F is Faraday's constant. For specific heat capacity, The specific enthalpy of the gas / liquid phase. The thermal conductivity of the material, The characteristic thickness along the z-direction of the anode and cathode thickness. For gas / liquid phase velocity, For electron current density, Electron conductivity, The ion current density, It is the ionic conductivity. For reference equilibrium potential, For reference quality score, The gas constant is... For ambient temperature, For fluid pressure, For fluid dynamic viscosity, For porous media permeability, For electron current density, For electron potential, This is the ionic potential. Gj is the Gibbs free energy of the electrochemical reaction. The number of electrons exchanged in the electrochemical reaction. For apparent mole fraction, For reference pressure, Let i be the stoichiometric coefficient of component i in electrochemical reaction j. The activation energy of the reaction. For reference current density, For reference temperature, The transport coefficient for the reduction reaction, Here, h is the transport coefficient for the oxidation reaction, t is time, p is the phase, and it is used to distinguish between the gas and liquid phases in gas-liquid two-phase flow phenomena. c Uj is the enthalpy term of convective heat transfer within the porous electrode, Uj is the half-cell potential of the electrochemical reaction, and Ij is the current density of the electrochemical reaction. Let be the local overpotential correlation factor for the j-th electrochemical reaction. This is the overpotential on the anode side. This represents the electron potential on the anode side. This represents the ionic potential on the anode side. The half-cell potential is on the anode side. This is the overpotential on the cathode side. This represents the half-cell potential on the cathode side. This represents the electron potential on the cathode side. This represents the ionic potential on the cathode side.

[0042] In the above embodiments of this application, ,unit ; ,unit ; ,unit ; ,unit ; The unit is m / s; ,unit F, unit ; ,unit ; ,unit ; ,unit ; The unit is meters (m). ,unit ; ,unit ; ,unit ; ,unit ; ,unit ; ,unit ; ,unit ; , ,unit ; , ,unit ; ,unit ; ,unit ; ,unit ; Gj, unit ; ,unit ; ,unit ; ,unit ; ,unit ; ,unit Uj, unit Ij, unit ; ,unit ; ,unit ; ,unit ; ,unit ; ,unit ; ,unit ; ,unit ; ,unit .

[0043] Regarding the construction process of the governing equations for the porous electrode model, specifically, along the thickness direction of the anode and cathode of the porous electrode, considering the matter, energy, charge, and flow characteristics at the microscopic pore scale, four types of conservation equations are integrated and combined with electrochemical reaction equations to complete the construction: 1. Construct conservation equations in modules: Mass conservation equation: Introducing the porosity of porous electrodes Volume fraction of gas and liquid phases Describes the mass fraction of the components. The time variation of the components is included, along with the diffusion flux, convective transport, and component generation / consumption due to electrochemical reactions: , By tracking the dynamic mass balance of various components within the pores at the microscopic level, this study reflects the consumption and generation of components by electrochemical reactions.

[0044] Energy conservation equation: taking into account the heat capacity of porous electrode solid framework ,density The temperature change, enthalpy change of the gas-liquid two phases, and the integration of heat conduction, convective heat transfer between gas, liquid and solid, Joule heating of electron and ion conduction, and heat effect of electrochemical reaction are used to characterize the heat transfer and temperature distribution of porous electrodes at the microscale.

[0045] The momentum conservation equation relates the pressure gradient along the thickness direction to the fluid viscous resistance, and combines it with the permeability K of the porous medium to describe the flow resistance and pressure changes of the gas and liquid phases in the micropores of the porous electrode, reflecting the constraint of the pore structure on fluid flow.

[0046] Electron / ion conservation equation: In the electron conservation equation, the electron current density The change along the thickness direction is equal to the electrochemical reaction formation rate. Combined with electronic conductivity The Ohmic relationship with the electron potential gradient describes electron transport in the solid framework; the ion conservation equation, on the other hand, describes the electron transport through the ion current density. Changes in electrochemical reaction rates, combined with ionic conductivity The ion potential gradient is used to characterize ion transport in porous electrolytes.

[0047] 2. Introducing the correlation and coupling of electrochemical reaction equations: The Nernst equation and the Butler-Volmer equation are introduced to characterize the thermodynamic and kinetic properties of electrochemical reactions. The electrochemical reaction process is linked to the above conservation equations in a closed loop. Finally, all equations are integrated to obtain the governing equations of the porous electrode model that accurately characterize the reaction and transport processes in the pores of the porous electrode at the microscopic level.

[0048] In particular, the Nernst equation is expressed by the mole fraction of the reacting components. Reference pressure Calculate the half-cell potential of an electrochemical reaction It reflects the influence of component concentration and temperature on the reaction equilibrium potential in porous electrodes at the microscale, accurately defines the thermodynamic equilibrium state of electrochemical reactions, and provides a benchmark for overpotential calculation.

[0049] The Butler-Volmer equation uses the activation energy of the reaction. Reference current density Using the core kinetic parameter, combined with the overpotential (calculated from the difference between the electronic / ionic potential and the equilibrium potential), the current density Ij of the electrochemical reaction is quantified, reflecting the kinetic rate of the electrochemical reaction within the micropores and directly describing the speed of the reaction.

[0050] The reaction current density Ij output by the Butler-Volmer equation corresponds to the electrochemical reaction formation rate in the electron and ion conservation equations. This becomes the core driving force of charge transport; at the same time The component source terms that will be substituted into the mass conservation equation will determine the generation / consumption rate of each component in the porous electrode; the thermal effect of the reaction will also be included in the energy conservation equation as a heat source term, ultimately achieving deep coupling between electrochemical reaction kinetic parameters (activation energy, reference current density, transport coefficient, etc.) and the mass, energy, and charge conservation equations, making the electrochemical reaction the core driver of all transport processes in the porous electrode.

[0051] Furthermore, in the full-scale coupled solution, the role of the governing equations for the porous electrode model is as follows: As the core reaction unit of the electrolyzer, the calculation results of the porous electrode model provide key boundaries for other scale models: transferring electron current density to the main electrode plate to provide input for its voltage loss and temperature calculation; transferring ion concentration and ion current to the diaphragm to support the calculation of mass transport and energy transfer of the diaphragm; receiving fluid temperature and component concentration from the flow channel model to realize convective heat transfer and component exchange between the flow channel and the porous electrode; and receiving conductivity parameter correction from the conductivity correction model, thus becoming the core hub of full-scale coupling.

[0052] When solving the core characteristics of the operating state simultaneously, the parameters output by the porous electrode model, such as the reaction current density, component concentration distribution, and temperature distribution, directly correspond to the core operating state of the electrolyzer: the reaction current density reflects the electrolysis yield, the component concentration reflects the consumption of reactants and the accumulation of products, and the temperature distribution reflects the reaction heat effect and heat dissipation capacity. These parameters are the core basis for judging the current energy efficiency and reaction efficiency of the electrolyzer.

[0053] The prediction of future operating status, combined with the real-time correction of reaction kinetic parameters by the HHDSDRGRA algorithm, allows the porous electrode model to simulate reaction dynamics under different operating parameters, and to deduce future reaction rates, component consumption, and temperature change trends. This provides core data support for the full-scale model to predict whether the electrolyzer will experience insufficient reactants, overheating, or decreased energy efficiency in the future, thus enabling the prediction of the long-term operating status of the electrolyzer.

[0054] Optionally, in step 104, the flow mass transfer parameters include the diffusion coefficient and the diffusion coefficient activation energy; the energy conservation equation in the governing equations of the diaphragm model is: ， The electron conservation equation is: , The ion conservation equation is: , The transmembrane transport mechanisms of various substances within the diaphragm are characterized by diaphragm diffusion calculation equations and diaphragm permeation calculation equations. The diaphragm diffusion calculation equation is as follows: , The equation for diaphragm permeability calculation is as follows: , The porosity of the porous electrode. For solid density, For the density of the liquid, For specific heat capacity, For ambient temperature, For liquid specific enthalpy, In the direction of membrane thickness, The ion current density, It is the ionic conductivity. For electron current density, Electron conductivity, For electric potential, Let i be the membrane diffusion coefficient of component i. Let i be the diffusion coefficient of component i at the reference temperature. The activation energy is the diffusion coefficient at the reference temperature. For reference temperature, Let i be the permeability coefficient of component i. Let i be the permeability coefficient of component i at the reference temperature. U is the activation energy of the permeability coefficient at the reference temperature, t is time, and U i This represents the ionic potential.

[0055] In the above embodiments of this application, ,unit ,unit ; ,unit ; ,unit ; ,unit , ,unit ; ,unit ; ,unit ; ,unit ; , ,unit ; , ,unit ; ,unit ; ,unit ; ,unit ; ,unit t, unit U i ,unit ; The electron conservation equation is also known as the electron current and potential equation, and the ion conservation equation is also known as the ion current and potential equation.

[0056] The process of constructing the governing equations for the diaphragm model involves expansion along the membrane thickness direction z. Combining three types of conservation equations and transmembrane transport mechanisms, the mass transport and energy transfer within the diaphragm are characterized in the mesoscopic dimension. The specific construction logic is as follows: 1. Dimensional expansion of the core conservation equation: Energy conservation equation: Left side of the equation: , Characterizing the solid skeleton within the diaphragm With liquid pores The amount of energy accumulated over time; Right side: , This is the thermal conductivity term along the thickness direction; , It is the Joule heat term generated by electron and ion transport. This equation fully covers the dynamic energy balance of heat accumulation, heat conduction and electrothermal generation in the membrane at the mesoscale.

[0057] Electron conservation equation: in Ohm's law form: , Characterizing electron transport along the thickness of the membrane, where electron conductivity... With electron potential gradient: , Determines electron current density This equation simultaneously reflects the core characteristics of the membrane—"blocking electrons and allowing ions to pass"—and quantifies the mesoscopic behavior of electron leakage.

[0058] Ion conservation equation: also in the form of Ohm's law: , Characterizing ion transport along the thickness direction, ionic conductivity With ion potential gradient: , Determines ion current density This is a mesoscopic quantitative characterization of the core function of the membrane: ion transmembrane conduction.

[0059] 2. Quantitative integration of transmembrane transport mechanisms: Transmembrane transport of substances within the membrane is characterized by two mechanisms: diffusion and osmosis. A temperature-dependent equation is introduced to correct the transport coefficient. The diffusion calculation equation uses a reference temperature Diffusion coefficient at the following levels With diffusion activation energy The effect of temperature on the diffusion capacity of components was quantified, and the diffusion behavior of components across membranes in the mesoscopic scale was characterized.

[0060] The permeability calculation equation, similarly, uses the reference permeability coefficient. To be compatible with osmotic activation energy The effect of temperature on the permeability of components is quantified, covering the permeation and transmembrane behavior of solvents and components.

[0061] By integrating the above three types of conservation equations with the transmembrane transport equations along the thickness direction, a mesoscopic membrane model governing equation is finally constructed. This equation focuses on the average characteristics of the overall material (ions, components) and energy transfer of the membrane, while also taking into account the influence of the internal solid-liquid two phases and temperature on the transport process.

[0062] Furthermore, in the process of solving the dynamic model of the electrolyzer and characterizing its operating state, the diaphragm model plays a crucial role as the core hub: The core bridge diaphragm of the multi-component coupling is a key component connecting the porous electrodes on both sides, and its calculated ion current density... Ion potential U i This serves as a boundary condition, passed to the porous electrode models on both sides, providing input for the electrochemical reactions and ion transport of the porous electrodes; simultaneously, it receives the temperature and component concentration parameters of the porous electrodes, realizing the material and energy interaction between the electrode and the membrane. Furthermore, the electronic current of the membrane... It will feed back to the main electrode model to correct the voltage loss calculation of the main electrode; the solution temperature and component concentration of the flow channel model will also affect the diffusion / permeability coefficient of the membrane through the transmembrane transport equation, realizing full-scale material-energy-charge coupling.

[0063] After solving the key characteristics of the current operating state simultaneously, the parameters output by the diaphragm model directly correspond to the core operating state of the electrolyzer: Ion current density This reflects the ion conduction efficiency of the electrolyzer, which corresponds to the current efficiency of the electrolyzer. The temperature distribution inside the diaphragm reflects the thermal equilibrium state and can be used to determine whether there is a risk of Joule heat accumulation or thermal runaway. Changes in the diffusion / permeability coefficient can characterize the aging and fouling of the membrane, and directly reflect the health status of the electrolyzer.

[0064] The prediction of future operating conditions combines the HHDSDRGRA algorithm with the conductivity correction model. The diaphragm model can predict the ion transport capacity and heat transfer efficiency under different operating conditions (temperature, component concentration). For example, by using the temperature dependence characteristics of the diffusion / permeation equation, it can predict the degradation of the diaphragm's transport performance under long-term high-temperature operation; by using the trend of ion current changes, it can predict the future current efficiency decline of the electrolyzer, providing a basis for the operation and maintenance and operating condition optimization of the electrolyzer.

[0065] Optionally, in step 105, the governing equations of the conductivity micro-correction model include the electronic conductivity calculation equation and the ionic conductivity calculation equation. The equation for calculating electronic conductivity is: , The equation for calculating ionic conductivity is: , Electron conductivity, Electronic conductivity at reference temperature The activation energy of electronic conductivity at the reference temperature. For ambient temperature, For reference temperature, This refers to ionic conductivity. For constant terms, The KOH concentration is a factor affecting the concentration. This represents the mole fraction of liquid water. η is the activation energy for ionic conductivity at the reference temperature.

[0066] In the above embodiments of this application, ,unit ; ,unit ; ,unit ; ,unit ; ,unit ; ,unit , ,unit ; ,unit .

[0067] The governing equations of the conductivity micro-correction model achieve precise micro-level correction of conductivity by considering ambient temperature, ion concentration, and solution temperature through multi-factor synergistic correction of electronic conductivity and ionic conductivity. The specific construction process is as follows: 1. Temperature-coordinated correction for electronic conductivity: The electronic conductivity calculation equation focuses on the influence of ambient temperature. Reference temperature Reference electronic conductivity Based on this, parameters related to the activation energy of electronic conductivity are used. The change in electronic conductivity when the ambient temperature T deviates from the reference temperature is quantified by an exponential function. When the ambient temperature rises, the thermal motion of electrons intensifies, and the electronic conductivity will be dynamically corrected according to an exponential law, thus completing the microscopic correction of electronic conductivity by the ambient temperature.

[0068] 2. Multi-factor synergistic correction of ionic conductivity: The ionic conductivity calculation equation simultaneously covers the synergistic effects of ion concentration and solution / ambient temperature. Correction for ion concentration: through the influence factor of KOH concentration With the mole fraction of liquid water xH2O, use (1 xH2O) characterizes the molar percentage of the electrolyte, combined with a constant term. The higher the ion concentration, the greater the increase in ion conductivity, thus achieving a microscopic correction of ion conductivity by ion concentration.

[0069] Temperature correction: through parameters related to ionic conductivity and activation energy , with the exponent: , The effect of solution / ambient temperature on ion migration was quantified. As temperature increases, ion thermal motion accelerates and migration resistance decreases, leading to a dynamic correction in ionic conductivity. By combining correction terms related to ion concentration and temperature, a synergistic microscopic correction of ionic conductivity based on multiple factors was achieved.

[0070] Finally, the electronic conductivity correction equation and the ionic conductivity correction equation are integrated to construct a conductivity micro-correction model control equation that covers the synergistic effects of ambient temperature, ion concentration, and solution temperature. This accurately corrects the calculated conductivity value from the perspective of microscopic particle thermal motion and ion concentration distribution.

[0071] In the process of solving the dynamic model of the electrolyzer and characterizing its current and future operating states, this model, as the core parameter correction module of the full-scale coupling, plays the following roles: Corrected electronic conductivity of the model output It directly inputs the electron conservation equations of the main electrode model and the diaphragm model, accurately correcting the calculation of electron current density; the corrected ionic conductivity... It will input the ion conservation equations of porous electrode models and membrane models, correct the core parameters of ion migration and transmembrane transport, avoid the calculation errors caused by fixed conductivity, and provide basic support for the accurate calculation of each component model.

[0072] Real-time linkage with operating status and dynamic correction of model accuracy: When the real-time operating parameters of the electrolyzer (ambient temperature, ion concentration, solution temperature) change, the model will output synchronously corrected conductivity parameters and feed them back to the fully coupled dynamic model. For example, when the solution temperature rises, the corrected ionic conductivity increases, which makes the ion current density calculated by the diaphragm model more realistic. This, in turn, makes the calculation of the electrochemical reaction rate of the porous electrode model and the voltage loss of the main electrode model more accurate, and ultimately reflects the accurate output of operating status parameters such as electrolyzer current efficiency and cell voltage.

[0073] When solving the current operating status: the conductivity is corrected based on real-time operating parameters, so that the calculation of each component model is more in line with the actual working conditions. The obtained state parameters such as cell voltage, current efficiency, and internal temperature distribution are more accurate and can truly reflect the current operating status of the electrolytic cell.

[0074] When predicting future operating conditions, the conductivity trend can be extrapolated in advance based on preset changes in operating conditions (such as increased ambient temperature and decreased ion concentration). This makes the future cell pressure, energy consumption, performance degradation, and other conditions predicted by the dynamic model more reliable, providing a basis for optimizing the operating conditions and making maintenance decisions for the electrolytic cell.

[0075] Optionally, in step 106, a heuristic high-dimensional space variable sensitivity reduced-order gradient regression algorithm is introduced to correct the reaction kinetic parameters and flow mass transfer parameters in real time, including: Step 1061: Determine the reaction kinetic parameters and flow mass transfer parameters as initial correction parameters; Step 1062: Perform sensitivity analysis on the initial correction parameters using local sensitivity indices, and after excluding the initial correction parameters that are below the sensitivity threshold from the sensitivity analysis results, obtain the target correction parameters. Step 1063: Configure sensitivity weighting factors for the target correction parameters, and divide the target correction parameters into high-sensitivity parameters and general-sensitivity parameters based on the median. Step 1064: Using a heuristic algorithm based on particle swarm optimization, a certain proportion of target correction parameters are randomly selected to participate in each optimization iteration, and the selection proportion is gradually increased as the iteration progresses and the error converges, until all target correction parameters are covered. The proportion allocation includes at least 50% of the highly sensitive parameters. Step 1065: Solve the gradient of the target correction parameter in high-dimensional space for each iteration. Based on the simultaneous equation method, add convergence error to the equation containing the target correction parameter, and solve the mathematical gradient value of the convergence error of each equation with respect to the target correction parameter. Step 1066: Using gradient descent and conjugate gradient methods as the core, and taking the sum of the convergence error values ​​of the equation after being weighted by the sensitivity weight factor as the optimization objective, the objective correction parameters are optimized and iterated step by step until the weighted convergence error value of the equation is minimized.

[0076] In the above embodiments of this application, the overall process for optimizing and regressing reaction kinetic parameters and flow mass transfer parameters is as follows: Sensitivity analysis is performed on the reaction kinetic parameters and flow mass transfer parameters to be corrected (initial correction parameters). All initial correction parameters are sorted from highest to lowest sensitivity to KPI error. After excluding parameters below the sensitivity threshold, sensitivity weight factors are assigned to the remaining parameters (target correction parameters), and high-sensitivity parameters and general-sensitivity parameters are divided based on the median. This step enables preliminary screening of redundant parameters, highlighting the dominant role of high-impact parameters and laying the foundation for subsequent optimization iterations. Using a heuristic algorithm based on particle swarm optimization, a certain proportion of target correction parameters are randomly selected to participate in each optimization iteration. This proportion must include at least 50% of the high-sensitivity parameters, and the selection proportion is gradually increased as iteration progresses and error converges until all target correction parameters are covered. This heuristic screening strategy effectively reduces the optimization complexity in high-dimensional space while ensuring the priority optimization of high-sensitivity parameters, thus improving iteration efficiency. For each iteration, the gradient of the target correction parameter is calculated in high-dimensional space. This is based on the simultaneous equation method, adding convergence error to the equations containing the relevant target correction parameters, and solving for the mathematical gradient of the convergence error of each equation with respect to the relevant parameters, thus achieving reduced-order gradient calculation in high-dimensional variable space. This avoids the redundancy of full-variable gradient calculation in high-dimensional space, achieving dimensionality reduction optimization through local gradient solving and reducing computational costs. Using gradient descent and conjugate gradient methods as the core, and with the sum of the convergence errors of the equations weighted by sensitivity weights as the optimization objective, the target correction variable is optimized iteratively until the weighted convergence error of the equations is minimized. This step achieves accurate parameter correction through gradient optimization, and the sensitivity weights ensure that the optimization results conform to actual needs, ultimately achieving efficient and accurate regression of high-dimensional parameters.

[0077] The gradient optimization process requires calculation and processing based on real-time runtime data, as follows: 1. Extract equipment process operation data through ETL technology and store it for later use in a predetermined manner.

[0078] 2. Based on different calculation step sizes, retrieve the input data of each device in the specified time period as the model input value, and retrieve the key operating parameters of the device in the corresponding time period as KPI indicator values.

[0079] 3. Substitute the initial values ​​of all variables to be corrected and the model input values ​​and KPI values ​​obtained in the previous step into the system of equations. At this point, the relevant equations will generate errors.

[0080] 4. Calculate the errors of each related equation and the weighted total error. Use the perturbation method to calculate the gradient of the variable to be calibrated based on the error of the related equation, and use the conjugate descent direction of the gradient as the direction of parameter optimization.

[0081] 5. Adjust the value of the parameter to be calibrated in a certain step size, and repeat steps 2 to 4.

[0082] 6. Monitor the changes in the weighted total error. When the weighted total error no longer changes within a certain number of steps, the gradient optimization calculation is complete.

[0083] Furthermore, when conducting sensitivity analysis, since there are many parameters to be corrected, and some empirically-based parameters have a wide range of values, a local sensitivity index can be used for evaluation. In practice, the absolute value of the sensitivity coefficient is calculated through local perturbations of the initial values. The absolute value of the coefficient will inevitably be in the range of 0 to 1, so 0.01 can be taken as the "sensitivity threshold". If the absolute value of the sensitivity coefficient of a parameter is less than 0.01, it is considered insensitive and excluded. The absolute value of the sensitivity coefficient is then used as the sensitivity weighting factor.

[0084] Furthermore, real-time correction is essentially a process of continuously adjusting parameters based on the index error until the index error converges. The formulas for calculating the sensitivity coefficient, index error, and median are all general formulas. The final output is the reaction kinetic parameters and flow mass transfer parameters after the KPI index error converges.

[0085] To ensure the real-time performance and accuracy of the parameters, automated calibration can be performed using actual process measurement data: First, process operation data of the electrolyzer is extracted based on ETL technology and stored for later use according to a predetermined method. Then, based on different calculation step sizes, the inlet data of each equipment during a specified time period is retrieved as the model input parameters, and the outlet data of each corresponding time period is used as the model target output parameters. For each set of input and output data, the parameter adaptive module is called to tune the reaction parameters, and the obtained set of single-point operating condition tuning values ​​is used as the candidate parameter space. Subsequently, for all input and output parameters, the parameter adaptive module is called again within the candidate parameter space to perform global parameter optimization. The above data retrieval, single-point tuning, and global optimization process is repeated until the number of iterations reaches the upper limit or the model output parameter error meets the standard, ultimately achieving real-time and accurate correction of the reaction kinetic parameters.

[0086] The corrected reaction kinetic parameters will be synchronized to the electrochemical reaction correlation terms of the governing equations of the porous electrode model, providing accurate kinetic support for the full-scale coupling of the flow channel, main electrode plate, diaphragm and conductivity correction model. This ensures that when solving the electrolyzer dynamic model simultaneously, the actual reaction process can be more accurately reflected, and thus the current and future operating status parameters of the electrolyzer can be accurately output.

[0087] In particular, parameter optimization involves a significant amount of computation. One approach is to build an AI model based on existing computational data, using graph neural networks and physics-driven methods, and solve inverse problems based on physical constraints to achieve end-to-end automated parameter tuning. This corresponds to higher computational costs and longer modeling cycles. Another approach is to automate the calibration process using a powerful AI model implemented via an agent. This requires semantic alignment of error state judgments and also involves higher computational costs and longer modeling cycles.

[0088] The terms used in the above embodiments of this application are explained as follows: 1. Dynamic technology for the operating status of industrial equipment. Simulation technology is a simulation modeling technology that uses simulation hardware and software to reflect the behavior or process of a system through simulation experiments. This technology can be divided into two main categories: simulation methods for continuous systems and simulation methods for discrete event systems. This application pertains to the field of simulation methods for continuous systems.

[0089] 2. Hydrogen production through water electrolysis. This technology uses electrical energy to electrolyze water molecules into hydrogen and oxygen, and is a core technology in green hydrogen production systems. Mainstream water electrolysis technologies include alkaline electrolysis, proton exchange membrane electrolysis, and solid oxide electrolysis.

[0090] 3. Lumped parameter modeling technique. This technique ignores the spatially distributed parameter characteristics of the actual system and describes the system behavior through a finite number of state variables. The core mathematical model is an ordinary differential equation. In the engineering field, this technique is widely used to simplify the analysis of complex systems.

[0091] 4. Distributed Parametric Modeling (DPMA). This technique is a mathematical modeling method used to describe the continuous changes of system characteristics or dynamic behavior with spatial coordinates and time. Its core lies in the fact that the system state cannot be accurately described by a finite number of variables, but must be characterized by field functions (i.e., functions of spatial location). Therefore, its mathematical model is usually composed of partial differential equations (PDEs) or functional differential equations.

[0092] 5. Numerical Solution Techniques for Partial Differential Equations. This technique uses numerical computation methods to approximate solutions to partial differential equations on a computer. Its core methods include the finite difference method, the finite element method, and the finite volume method. By discretizing the original problem, it transforms it into a system of algebraic equations for solution, involving mathematical theories such as stability, convergence, and error estimation.

[0093] 6. Numerical Optimization Techniques. This technique refers to indirect optimization design methods that use mathematical methods to find the optimal solution. Its modern application is mainly characterized by computer numerical computation. Its core paradigm can be summarized as: under given constraints, finding the optimal solution (minimum or maximum value) of the objective function through a systematic approach.

[0094] 7. Time Series Database Technology. Time series databases are primarily used to process time-stamped data that is generated frequently, has a large volume of measurement data points, and is heavily dependent on the acquisition time. The core characteristics of this technology are high concurrency, large data capacity, high real-time response, and heavy write load but low read intensity.

[0095] 8. ETL Technology. ETL (Extraction, Transformation, Loading) describes the process of extracting, transforming, and loading data from a source to a destination. ETL is the process of extracting, cleaning, transforming, and formatting data of various forms and sources within an enterprise. Its purpose is to integrate scattered, disorganized, and inconsistent data within the enterprise so that it can be analyzed, processed, and applied subsequently.

[0096] Currently, traditional macroscopic modeling methods based on lumped parameter methods ignore too many structural details of equipment, failing to accurately capture the actual operating state of the equipment, resulting in generally low accuracy. While full 3D equipment simulation based on distributed modeling technology can provide detailed parameters of key internal equipment, it involves large computational loads, poor real-time performance, and is unsuitable for large-scale applications. Statistical models based solely on process operation data can achieve high data real-time performance, but require long-term operational data support, leading to long modeling cycles. Furthermore, due to the scarcity of fault data, they cannot effectively cover the equipment's operating state and are prone to outputting data that does not conform to physical reality.

[0097] To address this, the technical solution of this embodiment, including modeling, parameter estimation, and process data calibration, is applied as follows: First, the system is divided into micro-segments using a distributed parameter method, establishing energy conservation, mass conservation, momentum conservation, and electrochemical reaction mechanism models at multiple physical scales. These models include flow channel models, main electrode models, porous electrode models, and membrane models. Sub-models for gas-liquid separation and transmembrane transport are integrated, and the HHDSDRGRA algorithm is introduced to correct reaction kinetic parameters. Automated calibration is performed using measured process data to ensure minimal error under different load and temperature conditions. Modeling is based on a parameterized distributed model with sufficient expressive power, resulting in strong model interpretability. Simultaneously, modeling utilizes actual field operating data and physical constraints, resulting in a wide range of operating conditions and high accuracy. Prior knowledge is fully utilized to efficiently extract information from actual process operating data, keeping the modeling cycle and cost controllable. An adaptive parameter calculation module and automated calibration process are introduced, leading to high modeling efficiency. The overall system is easy to deploy, possesses strong versatility, higher accuracy, and better reliability. Multi-scale, multi-field coupled models can approximate actual field operation as closely as possible, ensuring both the versatility of the equipment model and the maximum modeling accuracy. Adaptive parameter correction methods can significantly reduce the modeling workload, greatly improving modeling efficiency while maintaining accuracy. Automated calibration methods can effectively utilize field data to obtain key modeling parameters, accurately capture the actual operating status of equipment on-site, and ensure stable long-term predictions.

[0098] Furthermore, as Figure 1 To specifically implement the method, this application provides a dynamic modeling system for electrolytic cells, such as... Figure 2 As shown, the system includes: The model building module 201 is used to integrate the mass conservation equation, energy conservation equation and momentum conservation equation of the fluid in the electrolyzer, expand along the flow direction of the fluid in the electrolyzer, and build the flow channel model control equation. The flow channel model control equation characterizes the gas-liquid two-phase flow phenomenon that occurs in the electrolyzer in the macroscopic dimension. The model building module 201 is also used to build a current conservation equation for quantifying the voltage loss of the main electrode plate in the electrolytic cell during the current transmission process, and to build an energy conservation equation characterizing the temperature change trend of the main electrode plate over time. The equation is expanded along the thickness direction of the anode and cathode in the main electrode plate, and the current conservation equation and the energy conservation equation are coupled to obtain the main electrode plate model control equation. The main electrode plate model control equation characterizes the main electrode plate in the mesoscopic dimension. The model building module 201 is also used to integrate the mass conservation equation, energy conservation equation, momentum conservation equation and electron conservation equation of each substance in the porous electrode in the electrolytic cell, and to construct the control equation of the porous electrode model. The control equation of the porous electrode model characterizes the porous electrode at the microscopic level. When constructing the control equation of the porous electrode model, the Nernst equation and the Butler-Volmer equation are introduced to characterize the electrochemical reaction process in the porous electrode, and the reaction kinetic parameters in the electrochemical reaction process are correlated with various conservation equations. The model building module 201 is also used to expand along the thickness direction of the membrane based on the energy conservation equation, electron conservation equation and ion conservation equation of the membrane in the electrolytic cell, as well as the transmembrane transport mechanism of each substance in the membrane, and to build the membrane model control equation. The membrane model control equation characterizes the material transport and energy transfer process in the membrane in the mesoscopic dimension. The model building module 201 is also used to consider the synergistic effects of ambient temperature, ion concentration and solution temperature on electronic conductivity during the operation of the electrolyzer, and to build the control equation of the conductivity micro-correction model. The multiphysics model fusion module 202 is used to introduce a heuristic high-dimensional space variable sensitivity reduced-order gradient regression algorithm to correct the reaction kinetic parameters and flow mass transfer parameters in real time. It also couples the control equations of the flow channel model, the main electrode model, the porous electrode model, the diaphragm model, and the conductivity correction model to construct the control equations of the electrolyzer dynamic model. Based on the real-time operating parameters of the electrolyzer, the current operating state and future operating state of the electrolyzer can be obtained by solving the control equations of the electrolyzer dynamic model. The operating state is represented by the parameter values ​​obtained by solving the control equations of the electrolyzer dynamic model.

[0099] It should be noted that other corresponding descriptions of the functional units involved in the electrolytic cell dynamic modeling system provided in this application embodiment can be found in the following references. Figure 1 The corresponding descriptions in the method will not be repeated here.

[0100] Based on the above, Figure 1 Accordingly, this application also provides a storage medium storing a computer program, which, when executed by a processor, implements the above-described method. Figure 1 The method for dynamic modeling of electrolytic cells is shown.

[0101] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as personal computer, server, or network device, etc.) to execute the methods described in various implementation scenarios of this application.

[0102] Based on the above, Figure 1 The method shown, and Figure 2 To achieve the above objectives, the virtual system embodiment shown in this application also provides a computer device, specifically a personal computer, server, network device, etc. This computer device includes a storage medium and a processor; the storage medium stores a computer program; the processor executes the computer program to achieve the above-described objectives. Figure 1 The method for dynamic modeling of electrolytic cells is shown.

[0103] Optionally, the computer device may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB ports, card reader ports, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Bluetooth interfaces, Wi-Fi interfaces), etc.

[0104] Those skilled in the art will understand that the computer device structure provided in this embodiment does not constitute a limitation on the computer device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0105] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages and stores the hardware and software resources of a computer device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software within the physical device.

[0106] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented using software plus necessary general-purpose hardware platforms, or it can be implemented in hardware to construct a full-scale dynamic model of the electrolyzer: establishing the gas-liquid two-phase flow control equation in the flow channel at the macroscopic level; establishing the main electrode current-thermal coupling and diaphragm mass-heat transfer control equations at the mesoscopic level; establishing the porous electrode multi-field coupling control equations (embedding Nernst and Butler-Volmer equations to correlate reaction kinetic parameters) at the microscopic level, and introducing a conductivity micro-correction model. The HHDSDRGRA algorithm is used to correct reaction kinetic parameters in real time, coupling the various models at the full scale, and solving the current and future operating states of the electrolyzer by combining real-time operating parameters. Focusing on the multi-field coupling characteristics of the electrolyzer's "mechanical-electrical-thermal-chemical" aspects, constructing a multi-scale high-fidelity model can improve the speed of predicting the state parameters of process equipment (electrolyzer), effectively supporting the efficient operation of the hydrogen production process system.

[0107] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application. Those skilled in the art will understand that the modules in the system of the embodiment scenario can be distributed throughout the system of the embodiment scenario as described, or they can be modified to reside in one or more systems different from this embodiment scenario. The modules of the above-described embodiment scenario can be combined into one module, or further divided into multiple sub-modules.

[0108] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of any particular implementation scenario. The above disclosures are merely a few specific implementation scenarios of this application; however, this application is not limited thereto, and any modifications that can be made by those skilled in the art should fall within the protection scope of this application.

Claims

1. A method for dynamic modeling of an electrolytic cell, characterized in that, The method includes: The mass conservation equation, energy conservation equation and momentum conservation equation of the fluid in the integrated electrolyzer are expanded along the flow direction of the fluid in the electrolyzer to construct the flow channel model control equation. The flow channel model control equation characterizes the gas-liquid two-phase flow phenomenon that occurs in the electrolyzer in the macroscopic dimension. A current conservation equation is constructed to quantify the voltage loss of the main electrode plate in the electrolytic cell during current transmission, and an energy conservation equation is constructed to characterize the temperature change trend of the main electrode plate over time. The equations are expanded along the thickness direction of the anode and cathode in the main electrode plate, and the current conservation equation and the energy conservation equation are coupled to obtain the main electrode plate model control equation. The main electrode plate model control equation characterizes the main electrode plate in the mesoscopic dimension. In the porous electrode of the integrated electrolytic cell, the mass conservation equation, energy conservation equation, momentum conservation equation, and electron conservation equation of each substance are used to construct the governing equation of the porous electrode model. The governing equation of the porous electrode model characterizes the porous electrode at the microscopic level. When constructing the governing equation of the porous electrode model, the Nernst equation and the Butler-Volmer equation are introduced to characterize the electrochemical reaction process in the porous electrode, and the reaction kinetic parameters in the electrochemical reaction process are correlated with various conservation equations. Based on the energy conservation equation, electron conservation equation, and ion conservation equation of the membrane in the electrolyzer, as well as the transmembrane transport mechanism of various substances in the membrane, the membrane model governing equation is constructed along the membrane thickness direction. The membrane model governing equation characterizes the material transport and energy transfer process in the membrane in the mesoscopic dimension. Considering the synergistic effects of ambient temperature, ion concentration, and solution temperature on electronic conductivity during electrolytic cell operation, a microscopic correction model governing equation for conductivity is constructed. A heuristic high-dimensional space variable sensitivity reduced-order gradient regression algorithm is introduced to correct reaction kinetic parameters and flow mass transfer parameters in real time. The control equations of the flow channel model, the main electrode model, the porous electrode model, the diaphragm model, and the conductivity correction model are coupled in a full-scale manner to construct the control equations of the electrolyzer dynamic model. Based on the real-time operating parameters of the electrolyzer, the current operating state and future operating state of the electrolyzer can be obtained by solving the control equations of the electrolyzer dynamic model. The operating state is characterized by the parameter values ​​obtained by solving the control equations of the electrolyzer dynamic model.

2. The method according to claim 1, characterized in that, The flow mass transfer parameters include the first and second flow resistance coefficients. The mass conservation equation in the flow channel model governing equations is: ， The energy conservation equation is: ， The momentum conservation equation is: ， Along the flow direction of the fluid in the electrolytic cell, This represents the volume fraction of the gas / liquid phase. The density of the gas / liquid phase. The mass fraction of component i in the gas / liquid phase. For the mass transfer source term of component i in the gas / liquid phase, For gas / liquid phase velocity, The enthalpy of the gas / liquid phase mass ratio. The heat flux density between the gas / liquid phase and the main electrode plate within the flow channel. The heat flux density between the gas / liquid phase and the porous electrode plate within the flow channel. For fluid pressure, For fluid velocity, For fluid dynamic viscosity, These are the first and second flow resistance coefficients, respectively. is the fluid density, t is time, and p is the phase, used to distinguish between the gas and liquid phases in gas-liquid two-phase flow phenomena.

3. The method according to claim 1, characterized in that, The current conservation equation in the governing equations of the main plate model is: ， The energy conservation equation is: ， The thickness is along the anode and cathode direction, and T is the ambient temperature. For specific heat capacity, For electron current density, Electron conductivity, The thermal conductivity of the main electrode material Let t be the electron potential, and t be time.

4. The method according to claim 1, characterized in that, The flow and mass transfer parameters include the permeability of the porous medium. In the governing equations of the porous electrode model, the electron conservation equations include the equations for electron current and potential, ion current and potential, and the equations for calculating overpotential. The mass conservation equation is: ， The energy conservation equation is: ， The momentum conservation equation is: ， The equations for electron current and potential are: ， ， The equations for ion current and potential are as follows: ， ， The equation for calculating overpotential is: ， ， The Nernst equation is: ， The Butler-Volmer equation is: ， The porosity of the porous electrode. This represents the volume fraction of the gas / liquid phase. The density of the gas / liquid phase. The mass fraction of component i in the gas / liquid phase. Let i be the molecular diffusion flux of component i. For the gas / liquid phase fluid velocity, In the direction of thickness of the anode and cathode, Let i be the stoichiometric coefficient of component i in the electrochemical reaction. The rate of electron generation in an electrochemical reaction. Where is the number of electrons transferred in the reaction, and F is Faraday's constant. For specific heat capacity, The specific enthalpy of the gas / liquid phase. The thermal conductivity of the material, The characteristic thickness along the z-direction of the anode and cathode thickness. For gas / liquid phase velocity, For electron current density, Electron conductivity, The ion current density, It is the ionic conductivity. For reference equilibrium potential, For reference quality score, The gas constant is... For ambient temperature, For fluid pressure, For fluid dynamic viscosity, For porous media permeability, For electron current density, For electron potential, This is the ionic potential. Gj is the Gibbs free energy of the electrochemical reaction. The number of electrons exchanged in the electrochemical reaction. For apparent mole fraction, For reference pressure, Let i be the stoichiometric coefficient of component i in electrochemical reaction j. The activation energy of the reaction. For reference current density, For reference temperature, The transport coefficient for the reduction reaction, Here, h is the transport coefficient for the oxidation reaction, t is time, p is the phase, and it is used to distinguish between the gas and liquid phases in gas-liquid two-phase flow phenomena. c Uj is the enthalpy term of convective heat transfer within the porous electrode, Uj is the half-cell potential of the electrochemical reaction, and Ij is the current density of the electrochemical reaction. Let be the local overpotential correlation factor for the j-th electrochemical reaction. This is the overpotential on the anode side. This represents the electron potential on the anode side. This represents the ionic potential on the anode side. The half-cell potential is on the anode side. This is the overpotential on the cathode side. This represents the half-cell potential on the cathode side. This represents the electron potential on the cathode side. This represents the ionic potential on the cathode side.

5. The method according to claim 1, characterized in that, The flow and mass transfer parameters include the diffusion coefficient and the activation energy of the diffusion coefficient; the energy conservation equation in the governing equations of the diaphragm model is: ， The electron conservation equation is: ， The ion conservation equation is: ， The transmembrane transport mechanisms of various substances within the diaphragm are characterized by diaphragm diffusion calculation equations and diaphragm permeation calculation equations. The diaphragm diffusion calculation equation is as follows: ， The equation for diaphragm permeability calculation is as follows: ， The porosity of the porous electrode. For solid density, For the density of the liquid, For specific heat capacity, For ambient temperature, For liquid specific enthalpy, In the direction of membrane thickness, The ion current density, It is the ionic conductivity. For electron current density, Electron conductivity, For electric potential, Let i be the membrane diffusion coefficient of component i. Let i be the diffusion coefficient of component i at the reference temperature. The activation energy is the diffusion coefficient at the reference temperature. For reference temperature, Let i be the permeability coefficient of component i. Let i be the permeability coefficient of component i at the reference temperature. U is the activation energy of the permeability coefficient at the reference temperature, t is time, and U i This represents the ionic potential.

6. The method according to claim 1, characterized in that, The governing equations for the microscopic correction model of electrical conductivity include equations for calculating electronic conductivity and equations for calculating ionic conductivity. The equation for calculating electronic conductivity is: ， The equation for calculating ionic conductivity is: ， Electron conductivity, Electronic conductivity at reference temperature The activation energy of electronic conductivity at the reference temperature. For ambient temperature, For reference temperature, This refers to ionic conductivity. For constant terms, The KOH concentration is a factor affecting the concentration. This represents the mole fraction of liquid water. η is the activation energy for ionic conductivity at the reference temperature.

7. The method according to claim 1, characterized in that, A heuristic high-dimensional space variable-sensitive reduced-order gradient regression algorithm is introduced to perform real-time correction of reaction kinetic parameters and flow mass transfer parameters, including: The reaction kinetics parameters and the flow mass transfer parameters were determined as the initial correction parameters; The initial correction parameters were subjected to sensitivity analysis using local sensitivity indices. After excluding the initial correction parameters that were below the sensitivity threshold from the sensitivity analysis results, the target correction parameters were obtained. Configure sensitivity weighting factors for the target correction parameters, and divide the target correction parameters into high-sensitivity parameters and general-sensitivity parameters based on the median; Using a heuristic algorithm based on particle swarm optimization, a certain proportion of target correction parameters are randomly selected to participate in each optimization iteration, and the selection proportion is gradually increased as the iteration progresses and the error converges, until all target correction parameters are covered. The proportion allocation includes at least 50% of the highly sensitive parameters. For each iteration of the target correction parameter, the gradient is solved in high-dimensional space. Based on the simultaneous equation method, convergence error is added to the equation containing the target correction parameter, and the mathematical gradient value of the convergence error of each equation with respect to the target correction parameter is solved. Using gradient descent and conjugate gradient methods as the core, and taking the sum of convergence error values ​​of the equation after being weighted by sensitivity weight factors as the optimization objective, the objective correction parameters are optimized and iterated step by step until the weighted convergence error value of the equation is minimized.

8. A dynamic modeling system for an electrolytic cell, characterized in that, The system includes: The model building module is used to integrate the mass conservation equation, energy conservation equation and momentum conservation equation of the fluid in the electrolyzer, expand along the flow direction of the fluid in the electrolyzer, and construct the flow channel model control equation. The flow channel model control equation characterizes the gas-liquid two-phase flow phenomenon that occurs in the electrolyzer in the macroscopic dimension. The model building module is also used to construct the current conservation equation for quantifying the voltage loss of the main electrode plate in the electrolytic cell during the current transmission process, and to construct the energy conservation equation characterizing the temperature change trend of the main electrode plate over time. The equation is expanded along the thickness direction of the anode and cathode in the main electrode plate, and the current conservation equation and the energy conservation equation are coupled to obtain the main electrode plate model control equation. The main electrode plate model control equation characterizes the main electrode plate in the mesoscopic dimension. The model building module is also used to synthesize the mass conservation equation, energy conservation equation, momentum conservation equation, and electron conservation equation of each substance in the porous electrode of the electrolyzer, and to construct the control equation of the porous electrode model. The control equation of the porous electrode model characterizes the porous electrode at the microscopic level. When constructing the control equation of the porous electrode model, the Nernst equation and the Butler-Volmer equation are introduced to characterize the electrochemical reaction process in the porous electrode, and the reaction kinetic parameters in the electrochemical reaction process are correlated with various conservation equations. The model building module is also used to construct the membrane model control equations based on the energy conservation equation, electron conservation equation and ion conservation equation of the membrane in the electrolyzer, as well as the transmembrane transport mechanism of each substance in the membrane, along the membrane thickness direction. The membrane model control equations characterize the material transport and energy transfer process in the membrane in the mesoscopic dimension. The model building module is also used to consider the synergistic effects of ambient temperature, ion concentration and solution temperature on electronic conductivity during the operation of the electrolyzer, and to build the control equations for the conductivity micro-correction model. The multiphysics model fusion module is used to introduce a heuristic high-dimensional spatial variable sensitivity reduced-order gradient regression algorithm to correct reaction kinetic parameters and flow mass transfer parameters in real time. It couples the control equations of the flow channel model, the main electrode model, the porous electrode model, the diaphragm model, and the conductivity correction model at all scales to construct the control equations of the electrolyzer dynamic model. Based on the real-time operating parameters of the electrolyzer, the current operating state and future operating state of the electrolyzer can be obtained by solving the control equations of the electrolyzer dynamic model. The operating state is represented by the parameter values ​​obtained by solving the control equations of the electrolyzer dynamic model.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method for dynamic modeling of an electrolytic cell as described in any one of claims 1 to 7.

10. A computer device, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for dynamic modeling of the electrolytic cell as described in any one of claims 1 to 7.

Images