Electrolytic cell stack analysis methods, apparatus, computer equipment and storage media

Abstract

This application relates to the field of electrochemical analysis technology, and in particular to methods, apparatus, computer equipment, and storage media for analyzing electrolytic battery stacks. The method includes: acquiring initial internal state data of the battery stack to be analyzed, including an initial temperature field, an initial gas concentration field, and an initial electrode damping coefficient; inputting the initial internal state data into a pre-constructed non-isothermal symplectic kinetic model to obtain the current density, updated temperature field, and updated concentration field of the battery stack; inputting the updated temperature field and current density into a dynamic equivalent circuit model to obtain the overpotential parameters of the battery stack; using the overpotential parameters, performing multi-scale coupled time integration based on the current density to obtain the thermal stress distribution of the battery stack; and generating the analysis results of the battery stack based on the current density, updated temperature field, updated concentration field, and thermal stress distribution, thereby improving the reliability of battery stack analysis.

Description

Technical Field

[0001] This application relates to the field of electrochemical analysis technology, and in particular to an analysis method, apparatus, computer equipment, and storage medium for electrolytic cell stacks. Background Technology

[0002] With the development of green energy, co-electrolysis battery stacks, as the core equipment for green hydrogen production, have attracted increasing attention for their performance. The performance of co-electrolysis battery stacks is usually significantly affected by the coupling effects of multiple physical fields such as electrolyte flow, electrochemical reaction, and heat conduction.

[0003] In related battery stack analysis techniques, regarding modeling accuracy, dynamic models of high-temperature co-electrolysis battery stacks generally adopt a step-by-step decoupling method, such as separating the finite element thermodynamic model from the electrochemical impedance spectroscopy. Regarding dynamic real-time performance, traditional symplectic algorithms cannot be effectively applied under non-isothermal conditions. Regarding parameter calibration, existing impedance spectroscopy feedback mechanisms rely on steady-state assumptions and fail to combine real-time concentration field and temperature field data for dynamic correction.

[0004] Therefore, during implementation, the relevant technologies still suffer from low reliability in analyzing electrolytic cell stacks. Summary of the Invention

[0005] Based on this, the purpose of this application is to at least solve one of the above-mentioned technical defects, especially the technical defect of low reliability in battery stack analysis in the prior art. This application provides an electrolytic battery stack analysis method, apparatus, computer equipment and storage medium.

[0006] In a first aspect, this application provides a method for analyzing electrolytic cell stacks, the method comprising:

[0007] Acquire the initial internal state data of the battery stack to be analyzed, which includes the initial temperature field, initial gas concentration field, and initial electrode damping coefficient.

[0008] The initial internal state data is input into the pre-built non-isothermal symplectic dynamic model, and the outputs are the current density, updated temperature field, and updated concentration field of the battery stack. The non-isothermal symplectic dynamic model includes the non-isothermal Hamiltonian of the heat exchange term, the non-isothermal Hamiltonian of the potential energy term, and the non-isothermal Hamiltonian of the kinetic energy term.

[0009] The updated temperature field and current density are input into a pre-built dynamic equivalent circuit model, and the overpotential parameters of the battery stack are output.

[0010] By utilizing the overpotential parameter and performing multi-scale coupled time integration based on current density, the thermal stress distribution of the battery stack is obtained.

[0011] The analysis results of the battery stack are generated based on the current density, updated temperature field, updated concentration field, and thermal stress distribution.

[0012] In one embodiment, the method further includes:

[0013] Acquire experimental observation data, initial activation energy data, and initial thermal conductivity data;

[0014] The updated temperature field, updated concentration field, and overpotential parameters are input into a pre-built observation data mapping model, and the output is the predicted observation state data.

[0015] The experimental observation data and the predicted observation state data are input into the pre-set loss value model, and the optimized activation energy and optimized thermal conductivity are output. The loss value model includes an activation energy term optimized from the initial activation energy data and a thermal conductivity term optimized from the initial thermal conductivity data.

[0016] Among them, the optimized activation energy is used to feed back to the potential energy term, and the optimized thermal conductivity is used to feed back to the multi-scale coupling time integral.

[0017] In one embodiment, the thermal stress distribution of the battery stack is obtained by using the overpotential parameter and performing multi-scale coupled time integration based on the current density, including:

[0018] The current density is updated using the overpotential parameter to obtain the updated current density;

[0019] Based on the updated current density and the preset fast-change time step, the Joule thermal power parameters of the fast-change time layer of the battery stack are obtained.

[0020] The global temperature of the slow-varying time layer is obtained based on the Joule thermal power parameters of the fast-varying time layer and the preset slow-varying time step.

[0021] The thermal stress distribution is obtained based on the global temperature, the preset reference temperature, and the preset Young's modulus.

[0022] In one embodiment, the updated temperature field and current density are input to a pre-built dynamic equivalent circuit model, and the overpotential parameters of the battery stack are output, including:

[0023] Obtain the actual operating potential applied externally to the battery stack;

[0024] The real-time thermodynamic potential is determined based on the updated concentration field, the updated temperature field, and the reference temperature; the concentrations of each gas corresponding to the updated concentration field are determined based on the current density in the dynamic equivalent circuit model.

[0025] By using a dynamic equivalent circuit model, the error between the actual operating potential and the real-time thermodynamic potential is iteratively adjusted to obtain the overpotential parameter.

[0026] In one embodiment, determining the real-time thermodynamic potential based on the updated concentration field, the updated temperature field, and the reference temperature includes:

[0027] The thermodynamic equilibrium potential of the battery stack is determined based on the updated concentration field and the updated temperature field.

[0028] Based on the updated temperature field and the reference temperature, the thermodynamic temperature-corrected potential is obtained;

[0029] The real-time thermodynamic potential is obtained by using the thermodynamic equilibrium potential and the thermodynamic temperature correction potential.

[0030] In one embodiment, after obtaining the global temperature of the slowly varying time layer based on the Joule thermal power parameters of the fast-changing time layer and a preset slow-changing time step, the method further includes:

[0031] The thermodynamic equilibrium potential and the thermodynamic temperature correction potential are updated using the global temperature.

[0032] In one embodiment, the non-isothermal symplectic kinetic model is constructed using a Hamiltonian, which is expressed as follows:

[0033]

[0034] in, For kinetic energy, It is a momentum vector. This is the quality matrix; It is the potential energy term; For heat exchange terms, The thermal conductivity coefficient, For gas temperature, This represents the initial temperature field.

[0035] Secondly, this application provides an electrolytic cell stack analysis apparatus, the apparatus comprising:

[0036] The initial state acquisition module is used to acquire the initial internal state data of the battery stack to be analyzed. The initial internal state data of the battery stack includes the initial temperature field, the initial gas concentration field, and the initial electrode damping coefficient.

[0037] The internal state update module is used to input the initial internal state data into the pre-built non-isothermal symplectic dynamic model and output the current density, updated temperature field, and updated concentration field of the battery stack. The non-isothermal symplectic dynamic model includes the non-isothermal Hamiltonian of the heat exchange term, the non-isothermal Hamiltonian of the potential energy term, and the non-isothermal Hamiltonian of the kinetic energy term.

[0038] The overpotential determination module is used to input the updated temperature field and current density into the pre-built dynamic equivalent circuit model and output the overpotential parameters of the battery stack.

[0039] The multi-scale coupling module is used to obtain the thermal stress distribution of the battery stack by performing multi-scale coupling time integration based on the current density using the overpotential parameter.

[0040] The battery stack analysis module is used to generate analysis results for the battery stack based on current density, updated temperature field, updated concentration field, and thermal stress distribution.

[0041] Thirdly, this application provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0042] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0043] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0044] The electrolytic cell stack analysis method, apparatus, computer equipment, and storage medium provided in this application can effectively process internal state data of the cell stack in different dimensions through a non-isothermal symplectic dynamic model. For example, it can simultaneously process electro-thermal-chemical field coupling effects in different dimensions such as temperature field, gas concentration field, and electrode damping. Furthermore, it can combine a multi-scale coupling time integration method to solve the time scale differences between different time layers and use a dynamic equivalent circuit model to correct overpotential parameters in real time. This can improve the modeling accuracy, dynamic real-time performance, and parameter calibration accuracy of the cell stack analysis, thereby improving the reliability of the electrolytic cell stack analysis. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 A flowchart illustrating an electrolytic battery stack analysis method provided in this application embodiment;

[0047] Figure 2 A flowchart illustrating the steps following the determination of thermal stress distribution in a battery stack, provided as an embodiment of this application;

[0048] Figure 3 A flowchart illustrating the steps for obtaining the thermal stress distribution of a battery stack, provided as an embodiment of this application;

[0049] Figure 4 A flowchart illustrating the steps for obtaining overpotential parameters of a battery stack, provided in an embodiment of this application;

[0050] Figure 5 This is a schematic diagram of the structure of an electrolytic battery stack analysis device provided in an embodiment of this application;

[0051] Figure 6 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0052] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0053] As a core component of green hydrogen production, the performance of co-electrolysis battery stacks is affected by the coupling effects of multiple physics fields. Traditional analytical methods employ a step-by-step decoupling approach to handle the electro-thermal-chemical fields, resulting in the inability to reflect dynamic coupling effects in real time. At high temperatures, the timescales of rapidly changing and slowly changing time layers differ significantly, making single-time-step integration insufficient to simultaneously handle millisecond-level electrochemical reactions and second-level heat transfer processes, leading to errors in thermal stress calculations. Existing parameter calibration methods rely on steady-state assumptions and do not incorporate real-time concentration and temperature field data, resulting in insufficient accuracy for online correction of activation energy and charge transfer resistance.

[0054] Based on this, this application provides an analysis method, apparatus, computer device, and storage medium for electrolytic battery stacks. Through a non-isothermal symplectic kinetic model, it can effectively process internal state data of the battery stack in different dimensions. For example, it can simultaneously handle electro-thermal-chemical field coupling effects in different dimensions, such as temperature field, gas concentration field, and electrode damping. Furthermore, it can combine a multi-scale coupling time integration method to resolve time scale differences between different time layers, and can use a dynamic equivalent circuit model to correct overpotential parameters in real time. This improves the modeling accuracy, dynamic real-time performance, and parameter calibration accuracy of battery stack analysis, thereby enhancing the reliability of electrolytic battery stack analysis.

[0055] In one exemplary embodiment, Figure 1 This is a flowchart illustrating an electrolytic cell stack analysis method provided in an embodiment of this application, as shown below. Figure 1As shown, an analysis method for an electrolytic cell stack is provided. The method is illustrated using a terminal as an example. It is understood that this method can also be applied to a server, and to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. In this embodiment, the method includes the following steps S101 to S105. Wherein:

[0056] S101. Obtain the initial internal state data of the battery stack to be analyzed, wherein the initial internal state data of the battery stack includes the initial temperature field, the initial gas concentration field, and the initial electrode damping coefficient.

[0057] Electrolytic cell stacks refer to large electrochemical devices composed of multiple electrolytic cell units (individual devices that undergo electrochemical reactions) stacked in series or parallel. Examples include water electrolysis hydrogen production stacks, CO2 electrolysis stacks, and fuel cell stacks (operating in the opposite mode). The stack structure allows for higher total voltage or current output within a compact space.

[0058] Initial internal state data refers to the set of thermodynamic state parameters during the initial operation of an electrolytic cell stack. The initial temperature field refers to the temperature distribution data at the initial moment. The initial gas concentration field refers to the concentration distribution data of various reactant and product gases at the initial moment in locations such as flow channels and porous electrode pores within the stack. The initial electrode damping coefficient refers to a parameter reflecting the inherent energy dissipation characteristics of the electrode material during the electrochemical reaction process.

[0059] For example, the initial internal state data of the electrolytic cell stack can be transmitted to a terminal, allowing the terminal to perform further analysis and processing on the input initial internal state data. The initial temperature field can be acquired through a distributed temperature sensor array, forming a three-dimensional spatial temperature distribution matrix. The gas concentration field can be monitored in real time using a multi-channel gas sampling system to monitor the hydrogen and oxygen concentration gradients in each region. The initial electrode damping coefficient can be obtained from Ni-GDC impedance test data.

[0060] S102. Input the initial internal state data into the pre-constructed non-isothermal symplectic dynamic model, and output the current density, updated temperature field, and updated concentration field of the battery stack. The non-isothermal symplectic dynamic model includes the non-isothermal Hamiltonian of the heat exchange term, the non-isothermal Hamiltonian of the potential energy term, and the non-isothermal Hamiltonian of the kinetic energy term.

[0061] The non-isothermal symplectic kinetic model refers to a Hamiltonian system that includes heat exchange, potential, and kinetic energy terms. Specifically, it can be implemented using a symplectic solver to numerically integrate the Hamiltonian equations. The kinetic energy term describes the electrolysis rate, the potential energy term characterizes the concentration gradient, and the heat exchange term reflects the thermal conduction between the gas and the battery. Current density refers to the current intensity (A / m²) passing through a unit electrode area. Current density directly reflects the rate and intensity of the electrochemical reaction. The updated temperature field refers to the new temperature distribution considering the heat of electrochemical reaction (activation heat, reversible heat) and Joule heat, as well as the new temperature distribution after heat exchange. The updated concentration field refers to the new gas concentration distribution considering the consumption / generation and mass transfer of the reaction.

[0062] For example, the terminal or server can input the acquired / received initial state parameters into a pre-built non-isothermal symplectic kinetic model to simulate the core electrochemical processes (reactions, mass transfer) occurring inside the battery stack and their coupling evolution with the temperature field. For instance, the current density distribution can be obtained by solving the Hamiltonian canonical equation, while updating the initial temperature field and the initial concentration field.

[0063] S103. The updated temperature field and current density are input into the pre-built dynamic equivalent circuit model, and the overpotential parameters of the battery stack are output.

[0064] The dynamic equivalent circuit model refers to the circuit network that reflects polarization losses. Specifically, it can be implemented by using variable resistors to simulate charge transfer impedance and dynamically correcting the equilibrium potential using a real-time temperature field. The overpotential parameter can refer to the overpotential (η), which is the difference between the actual electrode potential and its thermodynamic equilibrium potential (e.g., determined by the Nernst equation).

[0065] For example, the terminal can use a pre-built dynamic equivalent circuit model to process and update the temperature field and current density. The dynamic equivalent circuit model can calculate the thermodynamic equilibrium potential based on the updated temperature field and determine the overpotential parameters by combining the applied working potential.

[0066] S104. Using the overpotential parameter, multi-scale coupled time integration is performed based on the current density to obtain the thermal stress distribution of the battery stack.

[0067] Multiscale coupled time integration refers to a numerical calculation method that handles different time steps and can be used to process systems involving coupling of different time scales and different physical fields (electrochemical-thermal-mechanical). Specifically, it can be achieved by using an explicit-implicit hybrid algorithm to handle the Joule thermal power of the fast-changing time layer and the global temperature of the slow-changing layer, with the fast-changing time step set to the millisecond level and the slow-changing time step set to the second level.

[0068] Thermal stress distribution refers to the spatial distribution of mechanical stresses (tensile stress, compressive stress, and shear stress) generated within the materials due to uneven temperature distribution and differences in thermal expansion coefficients between materials within the battery stack. Thermal stress is one of the main causes of deformation, cracking, and failure of battery stack components (such as bipolar plates, seals, and membrane electrodes).

[0069] For example, when the terminal calculates the Joule thermal power using the fast-changing time layer, it can use the explicit Euler method for millisecond-level iteration; when it calculates the global temperature using the slow-changing time layer, it can use the implicit difference method for second-level integration to obtain the thermal stress distribution results of the battery stack. In this way, it can provide an analytical data basis for the analysis of the battery stack.

[0070] S105. Based on current density, updated temperature field, updated concentration field, and thermal stress distribution, the analysis results of the battery stack are generated.

[0071] The analysis results refer to the assessment conclusions on the performance, state, and reliability of the battery stack, which integrates key information from the calculations of all models. For example, it may include overall or local performance (voltage, efficiency, current distribution uniformity), the location and size of temperature hotspots, the magnitude and distribution of thermal stress, and predictions of high-stress / high-risk areas.

[0072] For example, the terminal can form a dynamically optimized electrolytic battery stack parameter system based on the battery stack analysis data determined in steps S101-S104, so that the analysis results of the electrolytic battery stack can be output through the parameter system.

[0073] In this embodiment, a non-isothermal symplectic kinetic model is employed to achieve real-time coupled calculation of the electro-thermal-chemical fields, overcoming the problem of asynchronous updates between the temperature and concentration fields caused by traditional step-by-step decoupling methods. Furthermore, by correlating the dynamic equivalent circuit model with real-time temperature field data, the accuracy of overpotential parameter calculation can be improved. A multi-scale coupling time integration strategy is also used to achieve cross-scale transfer of Joule thermal power and global temperature, thus enabling more accurate thermal stress calculations. This, in turn, improves the reliability of the analysis of electrolytic cell stacks.

[0074] In an exemplary embodiment, the non-isothermal symplectic kinetic model is constructed using a Hamiltonian, which is expressed as follows:

[0075]

[0076] in, For kinetic energy, It is a momentum vector. This is the quality matrix; It is the potential energy term; For heat exchange terms, The thermal conductivity coefficient, For gas temperature, This represents the initial temperature field.

[0077] As an example, obtain the initial internal state data of the battery stack to be analyzed. Specifically, this may include the initial temperature field. Initial gas concentration field and The initial electrode damping coefficient γ is also known as γ. γ can be obtained from Ni-GDC (Nickel-Gadolinium Doped Ceria) impedance measurement data.

[0078] Alternatively, the non-isothermal symplectic kinetic model is expressed as follows (1):

[0079] (1);

[0080] in, A Hamiltonian can specifically include the following:

[0081] The kinetic energy term characterizes the electrolysis reaction rate, and the momentum vector is... With the mass matrix The product of the inverse matrices reflects the energy distribution.

[0082] This is a potential energy term that can describe the gas concentration gradient ( The interaction between the electric potential field and the free energy reflects the free energy in thermodynamic equilibrium.

[0083] This is a heat exchange term, representing the driving force of the temperature difference between the gas and the battery stack. The thermal conductivity coefficient, For gas temperature, This is the real-time battery temperature; it can be determined by the Joule heat increment. renew.

[0084] The current density can be calculated using the model shown in expression (1). Update the temperature field and updated concentration field At the same time, it can also output updated momentum fields. In symplectic algorithms, momentum These are endogenous variables of the system, and their initial values ​​can be determined by generalized coordinates. The initial state and physical constraints are derived naturally without requiring additional input.

[0085] In practical applications, the variables in expression (1) can be described by the following definitions:

[0086] It can describe the evolution of the concentration field / potential field over time. The diffusion rate is driven by momentum.

[0087] In the expression:

[0088] The concentration diffusion force is driven by the potential energy gradient; The gradient of Joule thermal power in the concentration / potential field characterizes the local heat source distribution; Joule heat power density; For electrode damping loss (which can be obtained from Ni-GDC impedance data); The damping coefficient characterizes the inhibitory effect of the electrode material on the reaction rate. For time step. To update the momentum field.

[0089] Specifically, the updated temperature field can also be determined using the following expression (2):

[0090] (2);

[0091] in, ; ; This refers to the temperature field from the previous step (which could be the initial temperature field). The Joule heat increment can be obtained from the momentum equation. calculate. The resistance is ohmic and can be derived from the impedance characteristics of the electrode material. This refers to the heat capacity of the battery stack, which is related to the specific heat capacity and mass of the material.

[0092] The current density can be determined by the following expression (3):

[0093] (3);

[0094] Expression (3) can be derived from the kinetic energy term of Hamiltonian H;

[0095] Therefore, through the specific implementation methods described above in practical applications, an updated concentration field can be output using a non-isothermal symplectic kinetic model. Update the momentum field Update the temperature field Current density .

[0096] Furthermore, the updated temperature field can be... and current density The overpotential parameter η of the battery stack is calculated by inputting it into the dynamic equivalent circuit model. Using the overpotential parameter η, based on the current density... Perform multi-scale coupled time integration. As an example, Joule thermal power can be calculated at rapidly varying time scales. Furthermore, it can calculate the global temperature at a slowly varying time level. Thus, the thermal stress distribution is obtained. .

[0097] Based on current density, updated temperature field, updated concentration field, and thermal stress distribution, analysis results of the battery stack can be generated. These results can include key performance indicators such as electrolysis efficiency and thermomechanical stress.

[0098] In one exemplary embodiment, Figure 2 This application provides a flowchart illustrating the steps following the process of obtaining the thermal stress distribution of a battery stack, as shown in the embodiments of this application. Figure 2 As shown, it is possible to Figure 1 Based on this, an exemplary description is given of the steps following the acquisition of the thermal stress distribution of the battery stack. Specifically, this method may further include steps S201 to S203, wherein:

[0099] S201. Obtain experimental observation data, initial activation energy data, and initial thermal conductivity data.

[0100] S202. The updated temperature field, updated concentration field, and overpotential parameters are input into the pre-built observation data mapping model, and the predicted observation state data is output.

[0101] S203. Input the experimental observation data and the predicted observation state data into the pre-set loss value model, and output the optimized activation energy and optimized thermal conductivity; wherein, the loss value model includes an activation energy term optimized from the initial activation energy data and a thermal conductivity term optimized from the initial thermal conductivity data.

[0102] Among them, the optimized activation energy is used to feed back to the potential energy term, and the optimized thermal conductivity is used to feed back to the multi-scale coupling time integral.

[0103] Among these, experimental observation data can refer to key performance indicators measured during battery stack operation (such as CO2 conversion rate and temperature gradient distribution). Initial activation energy can refer to the initial estimated value of the reaction energy barrier, which determines the electrochemical reaction rate. Initial thermal conductivity can refer to the initial value of the material's equivalent thermal conductivity, which affects the uniformity of the temperature field.

[0104] The observation data mapping model can refer to a neural network or multinomial regression model, which maps the state parameters (concentration field, temperature field, overpotential) output by the algorithm to the observation space. For example, the overpotential parameter can be correlated with the gas diffusion rate by constructing a linear mapping relationship between the temperature field and the concentration field.

[0105] The loss value model refers to a model designed to minimize the error between the predicted data and the experimental data. This model can be achieved by designing the activation energy term as the mean square error between the initial and measured activation energies, and by introducing regularization constraints into the thermal conductivity term. The parameter optimization process can employ the gradient descent algorithm, adjusting the activation energy and thermal conductivity parameters through backpropagation to minimize the error between the predicted and experimental data.

[0106] For example, during the operation of the battery stack, real-time gas concentration distribution data and infrared thermal imaging data are integrated into an experimental observation dataset and transmitted to a terminal for battery stack analysis. The terminal can input updated temperature and concentration field data into an observation data mapping model to generate predicted observational state data including current density distribution and heat flux density. By comparing the predicted data with the experimental data, the loss function simultaneously considers the chemical kinetic error of the activation energy parameter and the thermal conduction error of the thermal conductivity parameter.

[0107] Optionally, the optimized activation energy parameter can be fed back to the potential energy term of the non-isothermal symplectic kinetic model to correct the calculated value of the activation energy barrier of the electrode reaction; and the optimized thermal conductivity parameter can be synchronized to the multi-scale coupled integral module to adjust the thermal conductivity coefficient between the fast-changing time layer and the slow-changing time layer. Through this dynamic closed-loop correction mechanism, the calculation error of the thermodynamic potential of the model under high-temperature conditions can be reduced, thereby significantly improving the real-time performance and accuracy of thermal stress distribution prediction.

[0108] In this embodiment, a direct correlation is established between the model prediction results and actual experimental data by mapping observation data to a model. Through a loss function that includes initial activation energy and thermal conductivity terms, dynamic optimization of the activation energy and thermal conductivity parameters can be achieved, thereby calibrating the potential energy term and heat conduction process of the theoretical model. The optimized parameters are fed back to the corresponding calculation modules, thereby improving the accuracy of electrochemical reaction potential energy calculation and thermal stress distribution calculation, and ultimately enhancing the reliability of the battery stack analysis results.

[0109] In one exemplary embodiment, Figure 3 This application provides a schematic flowchart of a step for obtaining the thermal stress distribution of a battery stack, as shown in the embodiment of the present application. Figure 3 As shown, it is possible to Figure 1Based on this, the steps included in the analysis method of electrolytic battery stack are illustrated by example. Specifically, this method may also include steps S301 to S303. In step S104, the thermal stress distribution of the battery stack is obtained by using the overpotential parameter and performing multi-scale coupled time integration based on the current density. Specifically, steps S301 to S304 may be included, wherein:

[0110] S301. Update the current density using the overpotential parameter to obtain the updated current density;

[0111] S302. Based on the updated current density and the preset fast-change time step, obtain the Joule thermal power parameters of the fast-change time layer of the battery stack.

[0112] S303. Based on the Joule thermal power parameters of the fast-changing time layer and the preset slow-changing time step, obtain the global temperature of the slow-changing time layer.

[0113] S304. Based on the global temperature, the preset reference temperature, and the preset Young's modulus, the thermal stress distribution is obtained.

[0114] Among these, the fast-changing time step can refer to the time step of a fast-changing time layer. The Joule thermal power parameter can refer to the thermal power generated when current passes through the battery's internal resistance. The slow-changing time step can refer to the time step of a slow-changing time layer. The global temperature can refer to the overall temperature distribution of the battery stack (such as temperature values ​​at various locations) or the average temperature, reflecting the thermal state of the battery stack. The reference temperature can refer to a pre-set baseline temperature. Young's modulus can refer to the inherent mechanical parameter of a material, specifically the ratio of stress to strain within the material's elastic deformation range.

[0115] For example, dynamic updates to the current density can be achieved by real-time acquisition of electrode interface polarization data. Fast-changing time steps are set to milliseconds to capture Joule thermal transient responses, for example, using a 0.1-10 millisecond step size for local thermal power integration. Slow-changing time steps can be set to seconds to accommodate temperature diffusion characteristics, for example, using a 1-30 second step size for global temperature field iterative calculations. During the temperature gradient and thermal stress conversion process, the Young's modulus value can be dynamically adjusted based on the material's phase transition characteristics, for example, using piecewise linear interpolation within the electrolyte's solid-liquid phase transition temperature range.

[0116] Optionally, during the Joule thermal power calculation stage, high-frequency sampling of the current density can be performed using millisecond-level time steps, effectively capturing abrupt changes in current density caused by bubble formation and rupture on the electrode surface. During the global temperature calculation stage, time integration of the Joule thermal power can be performed using second-level time steps, avoiding numerical oscillations in the temperature field caused by high-frequency fluctuations. This segmented processing method allows the fast-changing time-layer data to be filtered and down-converted before being input into the slow-changing layer. That is, the instantaneous Joule heat accumulation from the fast-changing time-layer can be averaged across the slow-changing layer, achieving thermodynamic stability constraints and avoiding interference from short-timescale fluctuations on long-term temperature predictions. Finally, by comparing the spatial gradient distribution of the real-time temperature field and the reference temperature field, combined with the spatial variation characteristics of the Young's modulus parameter, the thermal stress concentration region at the electrolyte / electrode interface can be accurately calculated.

[0117] In this embodiment, by using different time steps to process the Joule thermal power and temperature field evolution respectively, the transient changes of local thermal effects are effectively captured, while ensuring the stability of the global temperature field calculation. The multi-scale coupling method improves the accuracy of thermal stress distribution calculation, thereby enhancing the reliability of the battery stack analysis results.

[0118] In one specific implementation, steps S301-S304 can be achieved by the following expression:

[0119] (4);

[0120] Represents the time step of a single fast change (like, The Joule heat power generated within the ). It is an ohmic resistor.

[0121] (5); Expression (5) represents the time step of the fast-change time layer. Joule heat accumulation Passed as input to the slow layer (like, This ensures thermodynamic stability constraints.

[0122] The Joule thermal power density of the rapidly varying time layer (which can be derived from the momentum equation output of expression (1)) is in units of W / m³. For slow-changing time steps (e.g., 1 second), and for fast-changing time steps... (For example, 1 millisecond) to form multi-scale coupling; For the number of time steps in the fast-changing time layer ( ).

[0123] Therefore, the instantaneous Joule heat accumulation of the fast-changing time layer can be averaged to the slow-changing layer, achieving thermodynamic stability constraints and avoiding interference from short-timescale fluctuations on long-term temperature prediction.

[0124] In practical applications, the global temperature of the slowly varying time layer can also be obtained through the above expressions (4) and (5). And the thermal stress distribution, wherein the thermal stress distribution can be expressed by the following expression (6):

[0125] (6);

[0126] in, Young's modulus; For reference temperature (e.g., initial temperature field); is the coefficient of thermal expansion.

[0127] In one exemplary embodiment, Figure 4 This application provides a schematic flowchart illustrating the steps for obtaining overpotential parameters of a battery stack through an embodiment of the present application. Figure 4 As shown, it is possible to Figure 1 Based on this, the steps included in the analysis method of electrolytic battery stack are illustrated by example. In step S103, the updated temperature field and current density are input into the pre-built dynamic equivalent circuit model, and the overpotential parameters of the battery stack are output, including steps S401 to S403, wherein:

[0128] S401. Obtain the actual operating potential applied externally to the battery stack;

[0129] S402. Determine the real-time thermodynamic potential based on the updated concentration field, updated temperature field, and reference temperature; the gas concentrations corresponding to the updated concentration field are determined based on the current density in the dynamic equivalent circuit model.

[0130] S403. Using a dynamic equivalent circuit model, the error between the actual working potential and the real-time thermodynamic potential is iteratively adjusted to obtain the overpotential parameter.

[0131] The actual operating potential refers to the measurable voltage that is directly applied to both ends of the battery stack by an external power source.

[0132] For example, the actual operating potential can be acquired in real time by a voltage sensor or an external circuit monitoring module, such as by setting a high-precision voltage probe between the positive and negative electrodes of the battery stack.

[0133] In the dynamic equivalent circuit model, the terminal can capture the actual operating potential of the battery stack in real time through external measuring devices. It can also calculate the thermodynamic equilibrium potential based on updated temperature and concentration field data output from the non-isothermal symplectic kinetic model. Simultaneously, the terminal can generate a temperature-corrected potential based on the difference between the current temperature field and the reference temperature. Furthermore, the real-time thermodynamic potential can be obtained by superimposing the thermodynamic equilibrium potential and the thermodynamic temperature-corrected potential. And the overpotential parameter can be obtained from the difference between the actual operating potential and the real-time thermodynamic potential.

[0134] Therefore, steps S401 to S403 can reduce the error of the overpotential parameter by dynamically integrating the influence of temperature field changes on electrochemical potential, thereby improving the calculation accuracy of Joule thermal power and thermal stress distribution in subsequent multi-scale coupled time integration, and thus improving the reliability of battery stack analysis results.

[0135] In this embodiment, the calculation accuracy of overpotential parameters is improved by utilizing the influence of dynamic changes in the temperature field on the thermodynamic potential. Furthermore, by obtaining the actual operating potential and combining it with the real-time thermodynamic potential that considers temperature effects, the true electrochemical state of the battery stack can be accurately reflected. This provides high-precision input parameters for subsequent multi-scale coupled time integration, thereby improving the reliability of battery stack analysis.

[0136] In an exemplary embodiment, step S402, determining the real-time thermodynamic potential based on the updated concentration field, the updated temperature field, and the reference temperature, may include:

[0137] The thermodynamic equilibrium potential of the battery stack is determined based on the updated concentration field and the updated temperature field.

[0138] Based on the updated temperature field and the reference temperature, the thermodynamic temperature-corrected potential is obtained;

[0139] The real-time thermodynamic potential is obtained by using the thermodynamic equilibrium potential and the thermodynamic temperature correction potential.

[0140] Thermodynamic equilibrium potential refers to the theoretical potential at which an electrochemical reaction is in equilibrium under specific temperature and concentration conditions. Thermodynamic temperature correction potential refers to the potential deviation caused by the operating temperature deviating from the reference temperature. Real-time thermodynamic potential refers to the theoretical potential reference value under actual operating conditions. The determination of the thermodynamic equilibrium potential can be based on updated concentration and temperature field data, specifically calculated using the Nernst equation combined with local gas concentration and temperature.

[0141] For example, the thermodynamic equilibrium potential can be determined by real-time acquisition of gas concentration distribution data and temperature distribution data, combined with the Nernst equation for the electrochemical reaction. For instance, the gas concentration term and temperature term in the Nernst equation can be dynamically corrected, whereby the concentration term can be determined based on the local concentration ratio of hydrogen to oxygen in the updated concentration field, and the temperature term can be determined based on the current temperature value of the updated temperature field.

[0142] The generation of the thermodynamic temperature correction potential can be achieved by introducing a preset reference temperature, such as the initial temperature of the battery stack or the standard operating temperature. The difference between the updated temperature field and the reference temperature is calculated and then linear or nonlinear compensation is performed in combination with the temperature coefficient, as shown in expressions (9) to (11).

[0143] During the execution of the dynamic equivalent circuit model, the calculation of the thermodynamic equilibrium potential can utilize the latest concentration and temperature field data in each iteration. For example, during the operation of an electrolytic cell stack, the gas concentration field changes in real time with the electrolyte flow and reaction process, while the temperature field fluctuates dynamically due to the Joule heating effect and heat conduction. The current thermodynamic equilibrium potential can be obtained by substituting the hydrogen and oxygen concentrations in the concentration field into the Nernst equation and superimposing the temperature influence term of the temperature field on the equilibrium potential. Simultaneously, the calculation of the temperature-corrected potential requires comparing the current temperature field with a reference temperature. For instance, when the temperature field shifts due to external thermal disturbances, a benchmark compensation is established using the reference temperature to eliminate the interference of transient temperature changes on the potential. For example, under high-temperature conditions, if a local increase in the temperature field causes the thermodynamic potential to deviate from the standard value, a reverse correction can be performed using the benchmark potential corresponding to the reference temperature, ensuring the stability of the real-time thermodynamic potential calculation results. The data processing in the two steps described above can be performed in parallel. The calculation frequency of the thermodynamic equilibrium potential is consistent with the iteration step size of the dynamic equivalent circuit model, and the calibration period of the temperature correction potential can be dynamically adjusted according to the rate of change of the temperature field. Furthermore, the sum of the thermodynamic equilibrium potential and the thermodynamic temperature correction potential can be calculated to obtain the real-time thermodynamic potential. By coupling the dynamic data of the concentration field and temperature field in real time, and achieving benchmark calibration based on the reference temperature for temperature compensation, the calculation accuracy of the overpotential parameter is ultimately ensured to meet the requirements of dynamic operating conditions.

[0144] In this embodiment, the thermodynamic equilibrium potential reflects the true state of the current concentration and temperature fields, avoiding the hysteresis error caused by traditional methods that rely solely on initial state data. Simultaneously, by introducing a reference temperature to calibrate the temperature correction potential, thermodynamic potential drift caused by transient changes in the temperature field can be eliminated. This improves the accuracy of thermodynamic potential parameters under dynamic operating conditions, thereby enhancing the overall reliability of the battery stack analysis model.

[0145] In some exemplary specific examples, the dynamic equivalent circuit model is expressed as follows (7):

[0146] (7);

[0147] in, It is the ohmic resistance, which characterizes the sum of the resistance of the electrode material, electrolyte, and contact resistance, and is linearly related to the current density. It is the charge transfer resistance, reflecting the resistance to electrochemical reaction kinetics (such as the activation energy of CO2 reduction and H2O electrolysis). The double-layer capacitance describes the charge storage effect at the electrode / electrolyte interface and is related to the electrode surface area and dielectric constant. The diffusion impedance is caused by gas diffusion (such as the transport of CO2 in the electrode pores), and its source can be the diffusion coefficient q in expression (1). (This can be derived from molecular dynamics simulation data); ω is the angular frequency, which can be the frequency of the excitation signal used in impedance spectroscopy testing.

[0148] By iteratively updating the parameters output by the non-isothermal symplectic kinetic model in S102, the impedance spectrum feedback can be adjusted in real time. The error caused by concentration polarization in the compensation expression (1) is compensated.

[0149] Furthermore, it can be expressed as follows:

[0150] (8);

[0151] (9);

[0152] (10);

[0153] (11);

[0154] in, To correct the previous charge transfer resistance, To correct the charge transfer resistance, To adapt the learning rate, dynamically adjust The update amplitude should be adjusted to avoid overshooting or slow convergence. It is the partial derivative of current density with respect to overpotential, reflecting the sensitivity of the reaction rate. The voltage perturbation is used to excite the impedance spectrum response. The voltage applied across the high-temperature co-electrolytic cell stack by an external power source, i.e. the actual operating potential, can be directly measured. This is the real-time thermodynamic potential. For thermodynamic equilibrium potential, The electric potential is corrected for thermodynamic temperature.

[0155] Where R is the standard electrode potential, and R is the gas constant. It can be determined by the temperature field output in expression (1), where n is the electron transfer number, F is the Faraday constant, and the gas concentration is... [CO] and [H2] can be determined by the generalized coordinate vector q. The concentration field q in expression (1) contains , CO and H2 are generated through a high-temperature co-electrolysis reaction. According to Faraday's law of electrolysis and the electrochemical reaction rate equation, the electrolysis reaction of CO2 and H2O generates CO and H2, as shown in expression (12):

[0156] (12);

[0157] The generation rate and current density of expression (12) Related, that is , .

[0158] The thermal correction term can be determined using expression (13). :

[0159] (13);

[0160] in, For temperature coefficient, This is a reference temperature (e.g., 25℃).

[0161] Therefore, based on expressions (8) to (13), the corrected charge transfer resistance can be output. ( The initial value can be measured by steady-state polarization experiments to determine the exchange current density. According to the formula Calculate initial values (Exchange current density), double-layer capacitance (Initial values ​​were obtained by fitting an equivalent circuit model through electrochemical impedance spectroscopy (EIS) experiments. In the equivalent circuit...) The capacitance characteristics at the corresponding electrode / electrolyte interface (its value is updated in real time through frequency domain fitting), and the adjusted overpotential parameters. .

[0162] In an exemplary embodiment, after obtaining the global temperature of the slowly varying time layer based on the Joule thermal power parameters of the fast-changing time layer and the preset slow-changing time step, the method further includes:

[0163] The thermodynamic equilibrium potential and the thermodynamic temperature correction potential are updated using the global temperature.

[0164] For example, the global temperature can be calculated using a slowly varying time step. The update of the thermodynamic equilibrium potential can be achieved through the nonlinear relationship between the concentration field and the temperature field. The update of the thermodynamic temperature correction potential can be achieved through the difference between the temperature gradient and the reference temperature. The updated thermodynamic equilibrium potential is used to adjust the driving force of the electrode reaction, and its change can be positively correlated with the temperature change. The updated thermodynamic temperature correction potential is used to compensate for the potential error caused by uneven temperature distribution; the compensation amount is dynamically adjusted according to the difference between the local temperature and the global temperature.

[0165] After obtaining the global temperature of the slowly varying time step, this global temperature can be synchronously input into the thermodynamic equilibrium potential calculation module. The thermodynamic equilibrium potential can be updated by correcting the temperature term in the model, for example, by replacing the initial temperature field in the model with the current global temperature value. Simultaneously, the thermodynamic temperature-corrected potential can be updated by comparing the difference between the current global temperature and the reference temperature. The updated potential parameters can be fed back to the dynamic equivalent circuit model in real time for recalculating the overpotential parameters. This allows the current density calculation to simultaneously include dynamic corrections for the influence of the temperature field on the thermodynamic equilibrium state and the temperature gradient, thereby eliminating potential calculation errors caused by temperature parameter lag. Through dual potential updates within each slowly varying time step, synchronous coupling of the temperature field and electrochemical parameters on a millisecond-level timescale is achieved.

[0166] In this embodiment, the battery stack model can more accurately reflect the impact of temperature changes on electrochemical reactions, improving the dynamic consistency of thermo-electric parameters during multi-scale coupling. Furthermore, it avoids the accumulated errors caused by the decoupling update of temperature field and potential parameters in traditional models, thereby enhancing the accuracy and reliability of battery stack performance prediction.

[0167] In one specific implementation, the specific implementation of S201 to S203 may include:

[0168] Based on expressions (1) to (13), the above output can be... , , Experimental observation data (CO2 conversion rate, temperature gradient), the loss value is determined by the following expression (14): (14);

[0169] in, For the model prediction function, the state parameters output by expressions (1) to (13) can be used. Mapped to the observation space, It can be a mapping input to the model. The observation error covariance matrix is ​​used to quantify the noise level of the experimental data. For initial model state parameters (such as activation energy) Thermal conductivity (This section needs optimization and adjustment.) These are background field parameters (prior estimates), which can come from historical data or physical models. Let be the background error covariance matrix, representing the prior uncertainty of the parameters.

[0170] In the first term of expression (14), the model aims to minimize the difference (residual) between the model prediction and the experimental data to ensure the physical consistency of the model. In the second term, the model aims to constrain the adjustment range of parameters to prevent excessive deviation from prior knowledge and ensure numerical stability.

[0171] Furthermore, the optimized activation energy can be determined by the following expression (15):

[0172] (15);

[0173] in, The activation energy before optimization. The learning rate (step size) is the rate at which the control parameter is adjusted, and it is usually adjusted adaptively. The gradient of the objective function with respect to the activation energy reflects the sensitivity of the parameters to model errors.

[0174] In practical applications, the gradient descent method can be used to backpropagate errors and optimize key physical parameters (such as activation energy). and thermal conductivity This makes the model more closely resemble the behavior of real systems.

[0175] Optimized activation energy This can directly affect the potential energy function in the non-isothermal symplectic dynamic modeling of expression (1). Adjusting the reaction kinetics path. Correcting the thermal conductivity. (Data from high-temperature thermophysical property tests) can be input into multi-scale coupled integral processing to optimize the simulation accuracy of the global temperature field and reduce thermal stress errors.

[0176] It can also be achieved through variational objective functions. By fusing model predictions and measured data, the potential gradient is backpropagated to expression (1). and expression (8) parameter.

[0177] In some specific embodiments, by linking non-isothermal dynamic modeling, impedance spectrum feedback, multi-scale thermodynamic integration, and data assimilation, a dynamically optimized electrolytic cell stack parameter system (including concentration field, temperature field, impedance characteristics, and material performance parameters) is finally output, which can achieve the following core functions:

[0178] Precise dynamic characteristic analysis: Real-time tracking of the multi-physical coupling effects of electro-thermal-chemical fields during high-temperature co-electrolysis improves the accuracy of transient response simulation. Enhanced stability: Impedance feedback and data assimilation compensate for model errors, suppressing performance degradation caused by concentration polarization or temperature gradients. Lifetime and efficiency optimization: Optimized activation energy parameters lower the reaction energy barrier, improving CO2 conversion efficiency; corrected thermal conductivity improves heat distribution uniformity, reducing thermal stress damage to materials. This application can provide a theoretical basis for the design and operation control of high-temperature electrolytic cell stacks. For example, dynamic parameter optimization can improve energy conversion efficiency (such as hydrogen / syngas yield), extend stack life (reduce thermal fatigue risk), and support stable operation under fluctuating renewable energy input scenarios.

[0179] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0180] The following describes the electrolytic battery stack analysis apparatus provided in the embodiments of this application. The electrolytic battery stack analysis apparatus has the same inventive concept as the electrolytic battery stack analysis method described above. The solution to the problem provided by the apparatus is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the electrolytic battery stack analysis apparatus provided below can be referred to the limitations of the electrolytic battery stack analysis method above. The electrolytic battery stack analysis apparatus described below and the electrolytic battery stack analysis method described above can be referred to each other, and will not be repeated here.

[0181] In one exemplary embodiment, Figure 5 This is a schematic diagram of the structure of an electrolytic cell stack analysis device provided in an embodiment of this application, as shown below. Figure 5As shown, the electrolytic cell stack analysis device 50 includes: an initial state acquisition module 510, an internal state update module 520, an overpotential determination module 530, a multi-scale coupling module 540, and a cell stack analysis module 550, wherein:

[0182] The initial state acquisition module 510 is used to acquire the initial internal state data of the battery stack to be analyzed, wherein the initial internal state data of the battery stack includes the initial temperature field, the initial gas concentration field, and the initial electrode damping coefficient.

[0183] The internal state update module 520 is used to input the initial internal state data into the pre-built non-isothermal symplectic dynamic model and output the current density, updated temperature field, and updated concentration field of the battery stack. The non-isothermal symplectic dynamic model includes the non-isothermal Hamiltonian of the heat exchange term, the non-isothermal Hamiltonian of the potential energy term, and the non-isothermal Hamiltonian of the kinetic energy term.

[0184] The overpotential determination module 530 is used to input the updated temperature field and current density into the pre-built dynamic equivalent circuit model and output the overpotential parameters of the battery stack.

[0185] The multi-scale coupling module 540 is used to obtain the thermal stress distribution of the battery stack by performing multi-scale coupling time integration based on the current density using the overpotential parameter.

[0186] The battery stack analysis module 550 is used to generate analysis results of the battery stack based on current density, updated temperature field, updated concentration field, and thermal stress distribution.

[0187] In one exemplary embodiment, the device further includes a sample and loss term module, an observation state prediction module, and a thermochemical optimization module.

[0188] The sample and loss term module is used to acquire experimental observation data, initial activation energy data, and initial thermal conductivity data.

[0189] The observation state prediction module is used to input the updated temperature field, updated concentration field, and overpotential parameters into a pre-built observation data mapping model, and output the observation state prediction data.

[0190] The thermochemical optimization module is used to input experimental observation data and observation state prediction data into a pre-set loss value model, and output the optimized activation energy and optimized thermal conductivity. The loss value model includes an activation energy term optimized from the initial activation energy data and a thermal conductivity term optimized from the initial thermal conductivity data.

[0191] Among them, the optimized activation energy is used to feed back to the potential energy term, and the optimized thermal conductivity is used to feed back to the multi-scale coupling time integral.

[0192] In an exemplary embodiment, the multi-scale coupling module 540 is used to update the current density using overpotential parameters to obtain an updated current density; based on the updated current density and a preset fast-change time step, the Joule thermal power parameters of the fast-change time layer of the battery stack are obtained; based on the Joule thermal power parameters of the fast-change time layer and a preset slow-change time step, the global temperature of the slow-change time layer is obtained; and based on the global temperature, a preset reference temperature, and a preset Young's modulus, the thermal stress distribution is obtained.

[0193] In an exemplary embodiment, the overpotential determination module 530 is used to obtain the actual operating potential applied to the battery stack by the external environment; determine the real-time thermodynamic potential based on the updated concentration field, the updated temperature field, and the reference temperature; the gas concentrations corresponding to the updated concentration field are determined based on the current density in the dynamic equivalent circuit model; and the error between the actual operating potential and the real-time thermodynamic potential is iteratively adjusted using the dynamic equivalent circuit model to obtain the overpotential parameters.

[0194] In an exemplary embodiment, the overpotential determination module 530 is used to determine the thermodynamic equilibrium potential of the battery stack based on the updated concentration field and the updated temperature field; to obtain the thermodynamic temperature correction potential based on the updated temperature field and the reference temperature; and to obtain the real-time thermodynamic potential based on the thermodynamic equilibrium potential and the thermodynamic temperature correction potential.

[0195] In one exemplary embodiment, the multi-scale coupling module 540 is used to update the thermodynamic equilibrium potential and the thermodynamic temperature correction potential using the global temperature.

[0196] In an exemplary embodiment, the non-isothermal symplectic kinetic model is constructed using a Hamiltonian, which is expressed as follows:

[0197]

[0198] in, For kinetic energy, It is a momentum vector. This is the quality matrix; It is the potential energy term; For heat exchange terms, The thermal conductivity coefficient, For gas temperature, This represents the initial temperature field.

[0199] In one exemplary embodiment, this application also provides a computer-readable storage medium storing a computer program that, when executed by one or more processors, causes the one or more processors to perform the steps of any of the electrolytic cell stack analysis methods described in the above embodiments.

[0200] In one exemplary embodiment, this application also provides a computer device storing a computer program in which computer-readable instructions, when executed by one or more processors, cause the one or more processors to perform the steps of any of the electrolytic cell stack analysis methods described in the above embodiments.

[0201] In one exemplary embodiment, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the electrolytic cell stack analysis methods described in the above embodiments.

[0202] Indicatively, such as Figure 6 As shown, Figure 6 This is a schematic diagram of the internal structure of a computer device 600 provided in an embodiment of this application. The computer device 600 can be provided as a server. (Refer to...) Figure 6 The computer device 600 includes a processing component 602, which further includes one or more processors, and memory resources represented by memory 601 for storing instructions, such as application programs, that can be executed by the processing component 602. The application programs stored in memory 601 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 602 is configured to execute instructions to perform the electrolytic cell stack analysis method of any of the above embodiments.

[0203] The computer device 600 may also include a power supply component 603 configured to perform power management of the computer device 600, a wired or wireless network interface 604 configured to connect the computer device 600 to a network, and an input / output (I / O) interface 605. The computer device 600 may operate on an operating system stored in memory 601, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0204] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0205] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0206] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0207] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for analyzing electrolytic cell stacks, characterized in that, The method includes: Acquire the initial internal state data of the battery stack to be analyzed, wherein the initial internal state data of the battery stack includes the initial temperature field, the initial gas concentration field, and the initial electrode damping coefficient; The initial internal state data is input into a pre-constructed non-isothermal symplectic dynamic model, which outputs the current density, updated temperature field, and updated concentration field of the battery stack. The non-isothermal symplectic dynamic model includes non-isothermal Hamiltonian terms for heat exchange, potential energy, and kinetic energy. The non-isothermal symplectic dynamic model is implemented by numerically integrating the Hamiltonian equation using a symplectic algorithm solver. The kinetic energy term describes the electrolysis rate, the potential energy term characterizes the concentration gradient, and the heat exchange term reflects the thermal conduction between the gas and the battery. The updated temperature field and current density are input into a pre-built dynamic equivalent circuit model, and the overpotential parameters of the battery stack are output. Using the overpotential parameter, multi-scale coupled time integration is performed based on the current density to obtain the thermal stress distribution of the battery stack. The analysis results of the battery stack are generated based on the current density, the updated temperature field, the updated concentration field, and the thermal stress distribution.

2. The method according to claim 1, characterized in that, The method further includes: Acquire experimental observation data, initial activation energy data, and initial thermal conductivity data; The updated temperature field, the updated concentration field, and the overpotential parameter are input into a pre-constructed observation data mapping model, and the observation state prediction data is output. The experimental observation data and the observation state prediction data are input into a pre-set loss value model, and the optimized activation energy and optimized thermal conductivity are output; wherein, the loss value model includes an activation energy term optimized from the initial activation energy data and a thermal conductivity term optimized from the initial thermal conductivity data; The optimized activation energy is used to feed back to the potential energy term, and the optimized thermal conductivity is used to feed back to the multi-scale coupling time integral.

3. The method according to claim 1, characterized in that, The step of using the overpotential parameter and performing multi-scale coupled time integration based on the current density to obtain the thermal stress distribution of the battery stack includes: The current density is updated using the overpotential parameter to obtain the updated current density; Based on the updated current density and the preset fast-change time step, the Joule thermal power parameters of the fast-change time layer of the battery stack are obtained. The global temperature of the slow-varying time layer is obtained based on the Joule thermal power parameters of the fast-varying time layer and the preset slow-varying time step. The thermal stress distribution is obtained based on the global temperature, the preset reference temperature, and the preset Young's modulus.

4. The method according to claim 3, characterized in that, The process of inputting the updated temperature field and current density into a pre-built dynamic equivalent circuit model and outputting the overpotential parameters of the battery stack includes: Obtain the actual operating potential applied externally to the battery stack; The real-time thermodynamic potential is determined based on the updated concentration field, the updated temperature field, and the reference temperature; the gas concentrations corresponding to the updated concentration field are determined based on the current density in the dynamic equivalent circuit model. By using a dynamic equivalent circuit model, the error between the actual operating potential and the real-time thermodynamic potential is iteratively adjusted to obtain the overpotential parameter.

5. The method according to claim 4, characterized in that, The step of determining the real-time thermodynamic potential based on the updated concentration field, the updated temperature field, and the reference temperature includes: Based on the updated concentration field and the updated temperature field, the thermodynamic equilibrium potential of the battery stack is determined; Based on the updated temperature field and the reference temperature, the thermodynamic temperature-corrected potential is obtained; The real-time thermodynamic potential is obtained based on the thermodynamic equilibrium potential and the thermodynamic temperature correction potential.

6. The method according to claim 5, characterized in that, After obtaining the global temperature of the slowly varying time layer based on the Joule thermal power parameters of the fast-changing time layer and the preset slow-changing time step, the method further includes: The thermodynamic equilibrium potential and the thermodynamic temperature correction potential are updated using the global temperature.

7. The method according to claim 1, characterized in that, The non-isothermal symplectic dynamic model is constructed using the Hamiltonian, which is expressed as follows: ； in, For kinetic energy, It is a momentum vector. This is the quality matrix; It is the potential energy term; For heat exchange terms, The thermal conductivity coefficient, For the gas temperature field, This represents the initial temperature field.

8. An electrolytic cell stack analysis apparatus, characterized in that, The device includes: The initial state acquisition module is used to acquire the initial internal state data of the battery stack to be analyzed, wherein the initial internal state data of the battery stack includes the initial temperature field, the initial gas concentration field, and the initial electrode damping coefficient. An internal state update module is used to input the initial internal state data into a pre-constructed non-isothermal symplectic dynamic model and output the current density, updated temperature field, and updated concentration field of the battery stack. The non-isothermal symplectic dynamic model includes non-isothermal Hamiltonian terms for heat exchange, potential energy, and kinetic energy. The non-isothermal symplectic dynamic model is implemented by numerically integrating the Hamiltonian equation using a symplectic algorithm solver. The kinetic energy term describes the electrolysis rate, the potential energy term characterizes the concentration gradient, and the heat exchange term reflects the thermal conduction between the gas and the battery. The overpotential determination module is used to input the updated temperature field and current density into a pre-built dynamic equivalent circuit model and output the overpotential parameters of the battery stack. A multi-scale coupling module is used to perform multi-scale coupling time integration based on the current density using the overpotential parameter to obtain the thermal stress distribution of the battery stack. The battery stack analysis module is used to generate analysis results for the battery stack based on the current density, the updated temperature field, the updated concentration field, and the thermal stress distribution.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images