Experimental verification method for multi-field intelligent simulation technology of water electrolyser

Abstract

The invention discloses an experimental verification method of a water electrolyser multi-field intelligent simulation technology, and relates to the technical field of water electrolyser simulation experiments. Comprising the steps that S1, a multi-physical field simulation model comprising a thermal field, a fluid field and at least one other physical field selected from an electric field, a chemical field or a structural field is constructed, dynamic coupling of the model is achieved through inter-field correlation parameters, and coupling of the thermal field and the fluid field comprises the steps that bubble generation kinetic parameters of the fluid field are dynamically corrected based on thermal field temperature distribution, and the correction adopts an Arrhenius equation to correlate the temperature and the bubble generation activation energy. The method has the advantages that temperature-bubble-heat transfer linkage is captured through a heat flow field dynamic coupling closed loop, multi-field optimization is driven, heat flow priority iteration and simulation experiments are used for synchronously improving efficiency in real time, convergence threshold values are dynamically adjusted, precision is guaranteed through full-dimensional verification, and accurate and reliable simulation support is provided for structural optimization and parameter debugging of the water electrolyser.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water electrolysis tank simulation experiment, in particular to a kind of water electrolysis tank multi-field intelligent simulation technology experimental verification method. BACKGROUND

[0002] As the core equipment of hydrogen energy production, the operation performance (such as hydrogen production efficiency, energy consumption, service life) of water electrolysis tank directly depends on the synergistic effect of thermal field, fluid field, electric field, chemical field and structural field. The existing water electrolysis tank multi-physical field simulation mostly adopts step-by-step coupling mode, which treats thermal field, fluid field and electric field, chemical field and the like equally, and ignores the core role of thermal-fluid coupling on the overall simulation results. For example, the temperature of thermal field is calculated first and then used as a fixed value in the fluid field, without realizing the real-time interaction of parameters between fields. Neither does it dynamically correct the bubble generation activation energy by temperature change (only uses fixed activation energy to calculate bubble rate), nor adjusts the heat transfer coefficient of thermal field according to bubble coverage rate, rising speed and other characteristics, resulting in significant deviation between simulation and experiment, large temperature deviation and large bubble coverage rate deviation, which cannot accurately reflect the actual linkage relationship of temperature-bubble-heat transfer in electrolysis tank. Therefore, we propose a kind of water electrolysis tank multi-field intelligent simulation technology experimental verification method to solve the above problems. SUMMARY

[0003] The purpose of the present application is to solve the problems raised in the background art by providing a kind of water electrolysis tank multi-field intelligent simulation technology experimental verification method.

[0004] In view of the above problems, the present application provides a kind of water electrolysis tank multi-field intelligent simulation technology experimental verification method, comprising the following steps:

[0005] S1, a multi-physical field simulation model containing thermal field, fluid field and at least one other physical field selected from electric field, chemical field or structural field is constructed, and the model is dynamically coupled through inter-field correlation parameters, wherein the coupling of thermal field and fluid field includes:

[0006] The bubble generation dynamics parameters of fluid field are dynamically corrected based on the temperature distribution of thermal field. The correction uses Arrhenius equation to correlate temperature and bubble generation activation energy. The activation energy Ea of the equation is expressed as Wherein Ea0 is the reference activation energy, and k is the temperature correction coefficient.

[0007] The heat transfer coefficient of thermal field is dynamically corrected based on the bubble characteristic parameters of fluid field. The bubble characteristic parameters include bubble coverage rate, rising speed and shape coefficient. The heat transfer coefficient h(φ) is expressed as Wherein φ is the bubble coverage rate, h0 is the reference heat transfer coefficient without bubble, and α is the coverage correction coefficient.

[0008] S2, correcting the inter-field correlation parameters by a bidirectional iterative calibration module, the iterative calibration focusing on the coupling of the thermal field and the fluid field, and the correlation parameters of other physical fields are corrected simultaneously, the bidirectional iterative calibration module preferentially performs the coupling iteration of the thermal field and the fluid field, and the iteration frequency is not less than twice the iteration frequency of the coupling of other physical fields;

[0009] S3, repeating step S2 until the deviation between the multi-field simulation result and the experimental data meets the preset convergence condition, the preset convergence condition including that the thermal field temperature deviation is less than ±2.5℃ and the fluid field bubble coverage deviation is less than ±10%, and obtaining the verified multi-field coupling simulation model.

[0010] Compared with the prior art, the technical scheme provided by the present application has at least the following technical effects or advantages:

[0011] 1: By establishing the real-time correlation between temperature and bubble generation activation energy, the thermal field temperature distribution dynamically corrects the bubble generation kinetic parameters of the fluid field, and simultaneously adjusts the heat transfer coefficient of the thermal field in the reverse direction according to the bubble coverage, rising speed and shape coefficient extracted from the fluid field, forming a real-time feedback closed loop to accurately capture the linkage relationship of temperature-bubble-heat transfer in the electrolytic cell, and driving the collaborative optimization of the electric field, chemical field and structural field. The prediction value truly reflects the running state of the electrolytic cell, and provides accurate simulation basis for the structural optimization and parameter debugging of the water electrolytic cell.

[0012] 2: The coupling iteration frequency of the thermal field and the fluid field is twice that of other physical fields, which preferentially solves the core deviation with the highest influence weight on the simulation accuracy, and then cooperatively calibrates the electric field, chemical field and structural field, avoids the interference of secondary field deviation on the adjustment of core parameters, and improves the iteration efficiency through the real-time synchronization of simulation software and experimental system and the systematic iterative calibration logic.

[0013] 3: The convergence threshold is dynamically adjusted according to the temperature gradient to ensure the simulation accuracy of high-risk areas, and the reliability of the model is ensured through the comparison and verification of experimental data and simulation results from temperature distribution, bubble characteristics to current density and electrode deformation. DETAILED DESCRIPTION

[0014] The above technical scheme will be described in detail below in combination with specific embodiments, so that the above technical scheme can be better understood. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments for explaining the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0015] A water electrolytic cell multi-field intelligent simulation technology experimental verification method comprises the following steps:

[0016] S1, constructing a multi-physical field simulation model comprising a thermal field, a fluid field, and at least one other physical field selected from an electric field, a chemical field, or a structural field, a three-dimensional geometric model can be established using SolidWorks. When setting, the size of the electrolytic cell can be set to 100 mm x 50 mm x 20 mm, and the electrode thickness can be set to 1 mm. The mesh can be divided using ANSYS Meshing. The thermal field and the fluid field use tetrahedral mesh (element number 1 x 10 5 ), and the electric field and the structural field use hexahedral mesh (element number 5 x 10 4 ). The mesh quality needs to satisfy the Jacobian > 0.7. The material properties are set according to experimental measurements, such as the elastic modulus of the electrode being 110 GPa, the Poisson's ratio being 0.3, the electrolyte conductivity being 5 S / m (25°C), and the electrolyte being prepared by placing the solution in a constant-temperature water bath at 25°C while stirring at a rate of 500 r / min for 15 minutes to ensure that KOH is completely dissolved and the concentration is uniform, avoiding local concentration differences that affect the accuracy of subsequent bubble generation and thermal field distribution simulations. The electrolyte conductivity can be measured using a DDS-307 conductivity meter, and the average value is 5 S / m when measured 5 times at 25°C.

[0017] It should be noted here that:

[0018] The thermal field uses the energy conservation equation, and the heat source term includes Joule heat and reaction heat, which can be collected by an infrared thermal imager. The infrared thermal imager is used to collect temperature field data, providing direct input for the coupling. The spatiotemporal variation of the temperature distribution will change T (temperature) in the Arrhenius equation in real time, thereby dynamically updating the bubble generation activation energy Ea, ensuring that the calculation of the bubble generation rate in the fluid field matches the actual temperature environment. When setting the infrared thermal imager, the lens is vertically aligned with the center of the top surface of the electrolytic cell, with a spacing of 50 cm. Before collection, the blackbody furnace (set to 25°C, 50°C, and 80°C as three standard temperature points) is used for calibration to ensure that the temperature measurement deviation in different regions does not exceed ±0.5°C. The conversion steps for the bubble generation rate are: application of Faraday's law → molar flow calculation → bubble number conversion, for example: calculate the molar flow of hydrogen gas according to Faraday's law: where i is the current density (A / m²), n = 2 (number of electron transfers), and F = 96485 C / mol; and calculate the bubble generation rate according to the bubble volume: where V m is the molar volume of gas (22.4 L / mol, standard state), and d is the bubble diameter (50 μm, measured by high-speed photography). Finally, the unit area bubble generation rate is obtained;

[0019] The fluid field is simulated using an Eulerian-Lagrange model to model gas-liquid two-phase flow. The bubble formation rate is calculated based on current density. High-speed photography is used to acquire bubble images, and parameters such as bubble coverage φ and morphology coefficient are extracted through image recognition and directly substituted into the heat transfer coefficient formula to correct the heat transfer efficiency calculation of "gas-liquid two-phase flow" in the thermal field (bubbles reduce the heat transfer capacity of the liquid; the higher the coverage, the smaller the heat transfer coefficient). This ensures that the thermal field model can reflect the influence of the actual flow field on heat transfer. The flow state of the electrolyte is calculated using the continuity equation (describing the fluid mass conservation) and the momentum equation (describing the fluid motion state, considering gravity and the force exerted by bubbles on the fluid). Bubbles are considered as discrete phases (Lagrange phases), and the generation, rise, and aggregation of bubbles are simulated by tracking the trajectory of individual bubbles. The bubble formation rate is based on current density data (following Faraday's law of electrolysis; the higher the current, the more gas is generated, and the faster the bubble formation rate). The bubble rise velocity and morphology coefficient are directly obtained from experimental data acquired by high-speed photography. The Schiller-Naumann model is also used. The drag force model describes the interaction force between the bubbles and the electrolyte (the bubbles are resisted by the electrolyte as they rise, and the magnitude of the resistance is related to the bubble speed and shape). When setting up, the center of the high-speed photography device lens can be aligned with the center of the side of the electrolytic cell at a distance of 30cm, and fixed with a tripod to avoid vibration. Before shooting, the correspondence between pixels and actual size is calibrated with a standard scale plate (minimum scale 0.01mm), and finally determined that 1 pixel corresponds to 0.05mm to ensure the measurement of parameters such as bubble diameter, rising speed, and shape coefficient.

[0020] The electric field can be described by the Poisson equation. The boundary conditions are that a voltage is applied to the anode and the cathode is grounded. The potential distribution can be acquired by a sensor array. The sensor array is used to acquire the current density. For example, the current density sensor uses 25 miniature platinum electrodes (distributed in a 5×5 matrix) and is attached to the cathode surface with a sensor spacing of 2mm. The current signal is acquired by an NI cDAQ-9178 data acquisition card with a sampling frequency of 100Hz to obtain the current distribution in different regions.

[0021] The chemical field uses the Butler-Volmer equation to calculate the reaction rate, which relies on the current density and temperature collected above. The current density data collected by the sensor array not only supports the calculation of the potential distribution of the electric field model, but also provides the "overpotential-current density" relationship for the Butler-Volmer equation of the chemical field, realizing the dynamic control of the reaction rate of the chemical field by the electric field.

[0022] The structural field uses the elasticity equation to calculate the electrode deformation, which can be collected by a sensor array. In this case, the sensor array is used to collect stress data. For example, eight Kistler 9311B piezoelectric sensors are selected as stress sensors and attached to the edge of the electrode (the area with the most obvious deformation). The measurement range is 0-50MPa, the resolution is 0.1MPa, and the sampling frequency is 50Hz. They are used to monitor the stress generated by the electrode due to temperature changes or current.

[0023] A more detailed explanation is as follows: The model achieves dynamic coupling through inter-field correlation parameters. To enable dynamic collaboration between the multi-physics simulation model and the experimental system, a bidirectional data transmission channel between the simulation software and the experimental acquisition system can be constructed via the Socket protocol. The specific transmission scheme is adapted to the model requirements of each physics field and the depth of the experimental acquisition logic. The transmitted data format must include timestamps, spatial coordinates, and corresponding core parameters of the physics field. The timestamps must be precisely synchronized with the sampling frequency of each acquisition device (e.g., matching the 100Hz temperature acquisition cycle of an infrared thermal imager and the 1000fps bubble image acquisition cycle of high-speed photography) to ensure the consistency of the time dimension of cross-field coupling parameters such as thermal field temperature distribution and fluid field bubble characteristics. The spatial coordinates must correspond to the physical region division of the water electrolyzer, so that the local data acquired in the experiment (e.g., current density and stress values ​​in specific areas) can accurately match the corresponding calculation units of the simulation model, supporting the boundary condition correction of the electric field Poisson equation and the local deformation calculation of the structural field elasticity equation; The field parameters need to cover the entire "input-output verification" chain of each field model, specifically including: temperature field data collected by infrared thermal imager (used to update the heat source term of the thermal field energy conservation equation in real time, and to dynamically correct the activation energy of bubble generation in the fluid field through the Arrhenius equation); bubble coverage and morphology coefficients extracted by high-speed photography (used to correct the heat transfer coefficient of the thermal field in real time and optimize the heat transfer efficiency calculation of "gas-liquid two-phase flow"); current density collected by sensor array (to simultaneously support the calculation of bubble generation rate in the fluid field, the back-inference of electric field potential distribution, and the construction of the "overpotential-current density" relationship of the Butler-Volmer equation in the chemical field); and stress data (used to verify the calculation results of electrode deformation in the structural field). At the same time, the data transmission delay is strictly controlled within 50ms to ensure that the dynamic parameters collected in the experiment (such as the spatiotemporal changes of the temperature field and the characteristics of bubble motion) can be fed back to the simulation model in real time, ensuring the timeliness of the coupling correction of the thermal field and the fluid field, as well as the synchronization of the dynamic coupling simulation results of each physical field with the experimental process.

[0024] It should also be noted that the coupling between the thermal field and the fluid field includes:

[0025] The bubble generation kinetic parameters of the fluid field are dynamically corrected based on the temperature distribution of the thermal field. This correction uses the Arrhenius equation to correlate temperature with the activation energy of bubble generation. The activation energy Ea in the equation is expressed as... As temperature T increases, the activation energy Ea(T) decreases linearly, with the decrease controlled by the coefficient k. Ea(T) is a temperature-dependent activation energy in kJ / mol or eV, dynamically changing with temperature T. Ea0 is the baseline activation energy, representing the initial value of the activation energy at a reference temperature (e.g., T=0), ranging from 40-50 kJ / mol. k is a temperature correction coefficient in kJ / (mol·K), quantifying the sensitivity of the activation energy to temperature changes. T represents absolute temperature (K, Kelvin). The Arrhenius equation is a classic formula for chemical kinetics, used to describe the effect of temperature on reaction rates or process kinetic parameters. Here, the activation energy is explicitly expressed as a linear function of temperature, and a temperature correction coefficient k is introduced. The temperature correction coefficient k is selected in the range of 0.008-0.012 kJ / (mol·℃), and the goodness of fit of the regression is required. The value should not be lower than 0.95 to ensure that the parameters match the actual bubble generation pattern.

[0026] The heat transfer coefficient of the thermal field is dynamically corrected based on the characteristic parameters of the bubble in the fluid field. These bubble characteristic parameters include bubble coverage (φ), rising velocity (v), and shape coefficient (ξ). The heat transfer coefficient h(φ) is expressed as... , where φ is the bubble coverage, h0 is the reference heat transfer coefficient without bubbles, and α is the coverage correction coefficient, which is calculated from the rise rate and the shape coefficient.

[0027] The data acquisition conditions for the bubble generation experiment covered the current density. For each working condition, data were collected at least three times for repeated experiments. The relative standard deviation of the experimental data was no more than 5%. The bubble generation experiment strictly followed the sequence of "temperature stabilization - current application - data acquisition": For a certain working condition, the electrolyte was first heated to the target temperature and the temperature was maintained stable for 10 minutes using a constant temperature water bath (to ensure uniform temperature in the tank). Then, the corresponding current was applied, and a high-speed photography device was started to record the bubble movement process for 30 seconds. After the images were captured, the images were processed using image analysis software (such as Image-Pro Plus) to extract the number and diameter of bubbles in each frame and calculate the bubble generation rate per unit time. The bubble generation rate and temperature data at different temperatures were imported into Origin software. Combining the principle of "temperature affects reaction activity" in chemical kinetics (the higher the temperature, the lower the activation energy required for the reaction, and the faster the bubble generation rate), two key parameters were determined by linear regression: the baseline activation energy (i.e., the initial activation energy at the reference temperature) and the temperature correction coefficient (quantifying the decrease in activation energy for every 1°C increase in temperature). Finally, the range of the baseline activation energy was determined to be 40-50 kJ / mol.

[0028] The range of values ​​for the reference heat transfer coefficient h0 mentioned in S1 is as follows: The coverage correction coefficient α ranges from 0.3 to 0.5. h0 and α are determined through heat flux density measurement experiments. The experiments use a miniature heat flux sensor array with a sensor spacing of no more than 2 mm and a response time of no more than 5 ms. The heat flux density measurement can use 16 Hukseflux HFP01 miniature heat flux sensors (distributed in a 4×4 matrix), with a sensor size of 5 mm × 5 mm, a spacing of 2 mm, and attached to the area of ​​the outer wall of the electrolytic cell in contact with the electrode. The response time is ≤ 5 ms, and the measurement range is 0-1000 W / m², which is used for subsequent calibration of the heat transfer coefficient.

[0029] The bubble characteristic parameters described in S1 are acquired synchronously by a high-speed photography device and an infrared thermal imager. The frame rate of the high-speed photography device is not less than 1000fps, the temperature measurement accuracy of the infrared thermal imager is not less than ±1℃, and the spatial resolution is not less than 640×512 pixels.

[0030] Specifically, the process is divided into two stages: bubble-free and bubble-containing. In the bubble-free stage (power to the electrolytic cell is turned off to prevent bubble formation), the electrolyte temperature is maintained at a constant temperature (e.g., 25°C). A heat flux sensor array is activated to collect heat flux density data over a certain period (e.g., 10 minutes), and the average value is taken as the baseline heat flux density for the bubble-free stage. Using Fourier's law of heat conduction (heat flux density equals the product of the heat transfer coefficient and the temperature difference), and given the known temperature difference between the electrolyte and the environment (measured by an infrared thermal imager), the baseline heat transfer coefficient for the bubble-free stage is calculated, with a value ranging from 700-900 W / (m²・K). In the bubble-containing stage, a two-step determination of the coverage correction coefficient (this coefficient describes the effect of bubble coverage on the heat transfer coefficient; more bubbles result in a lower heat transfer coefficient) is required: First, the bubble shape is fixed as spherical (achieved by fine-tuning the airflow generator, with the shape coefficient set to 1.0), and the bubble coverage is 20%. The bubble rising speed is changed (set to 0). The first step involves using five gradients (0.1 m / s, 0.2 m / s, 0.3 m / s, 0.4 m / s, and 0.5 m / s) to collect heat flux density at different velocities. The corresponding coverage correction coefficient is then derived, and linear regression is used to establish the relationship between the coverage correction coefficient and the rising speed, ultimately determining the speed influence coefficient to be 0.4 s / m. The second step fixes the bubble's rising speed at 0.3 m / s and the coverage at 20%, while changing the bubble shape coefficient (setting five gradients: 1.0, 0.9, 0.8, 0.7, and 0.6, with smaller coefficients indicating flatter bubbles). Again, heat flux data is used to deduce the coverage correction coefficient, and linear regression is used to establish the relationship between the coverage correction coefficient and the shape coefficient, determining the shape influence coefficient to be 0.3. Combining the results of these two steps, a dynamic calculation formula for the coverage correction coefficient is obtained, with a value range of 0.3-0.5, ensuring that this coefficient can be adjusted in real-time according to the actual movement of the bubble.

[0031] S2. The inter-field correlation parameters are corrected through a bidirectional iterative calibration module. This iterative calibration focuses on the coupling of the thermal and fluid fields, collaboratively correcting the correlation parameters of other physical fields. The correction of these parameters includes: correcting the chemical field reaction rate based on the electric field current density distribution. This reaction rate is calculated using the Butler-Volmer equation, where the overpotential η is determined based on the current density distribution i(x,y) and electrode material characteristics. These electrode material characteristics include the exchange current density and the transfer coefficient. The exchange current density is measured using linear sweep voltammetry (LSV) with a three-electrode system (working electrode: nickel electrode; reference electrode: Hg / HgO electrode; counter electrode: platinum electrode). The scan rate is 5 mV / s, and the potential range is 0-1.2 V vs Hg / HgO, obtained by fitting the slope of a Tafel curve. The complete calculation formula for the overpotential η is: Ohmic overpotential R is the electrolyte resistance (calculated from the conductivity σ and electrode geometry). (where d is the electrode spacing and S is the electrode area); other physical field-related parameter corrections also include: correcting the electric field electrode spacing based on the structural field stress distribution. This correction uses finite element analysis to convert the structural field stress distribution σ(x,y) into electrode deformation displacement Δd, thereby adjusting the electric field strength. Where d0 is the initial electrode spacing and U is the slot voltage, the electrode structure model is constructed using the SOLID186 element of ANSYS Mechanical, the stress distribution σ(x,y) is mapped to loads by nodes, and the Newton-Raphson algorithm is used to solve the electrode deformation displacement Δd. The displacement results are extracted through the Node Solution module.

[0032] The bidirectional iterative calibration module prioritizes the coupling iteration of the thermal field and the fluid field, with an iteration frequency no less than twice the coupling iteration frequency of other physical fields. The inter-field coupling order of the multiphysics simulation model is determined based on sensitivity analysis. For every two coupling iterations between the thermal field and the fluid field, one coupling iteration between other physical fields is completed. Inter-field coupling can be achieved through the System Coupling module of ANSYS Workbench, with the data transfer frequency synchronized with the iteration frequency. The data format is automatically converted using the software's built-in Data Transfer function (e.g., converting Fluent's .dat file to Mechanical's .rst file).

[0033] Two-way iterative calibration is a key step in improving simulation accuracy. It employs a "heat flux priority, layered iteration" strategy, prioritizing the simulation accuracy of the thermal and fluid fields before coordinating the calibration of other fields. This avoids deviations in secondary fields interfering with the adjustment of core parameters. The specific process is as follows:

[0034] Initialization settings before iteration: Input the previously calibrated baseline parameters (such as activation energy, heat transfer coefficient, and coverage correction coefficient) into the simulation model, set the iteration counter (initial value is 0), the maximum number of iterations (50 times to avoid infinite iteration), and the initial convergence threshold (thermal field temperature deviation is less than ±2.5℃, fluid field bubble coverage deviation is less than ±10%), and start the Socket data transmission channel to ensure that the simulation and experimental systems run synchronously.

[0035] The first step is the priority iteration of the thermal and fluid fields: In the first iteration (counter=1), the simulation model independently calculates the initial distribution of the thermal and fluid fields. The temperature distribution data of the thermal field (including the temperature value of each calculation unit, the corresponding timestamp, and spatial coordinates) is transmitted to the thermal field simulation software and the fluid field simulation software in real time via the Socket protocol, with a data update delay of no more than 50ms. After receiving the data (the data format includes timestamps, spatial coordinates, and corresponding physical field parameters, supporting breakpoint resume and data verification), the experimental system starts the infrared thermal imager to collect the actual temperature data at the corresponding timestamp and spatial location, and calculates the deviation between the simulated temperature and the experimental temperature. At the same time, the fluid field transmits the bubble coverage data to the experimental system, and the high-speed photography device collects the actual bubble coverage and calculates the deviation between the two. If the temperature deviation exceeds ±2.5℃, adjust the heat source term of the thermal field according to the direction of the deviation: if the simulated temperature is higher than the experimental temperature, it indicates that the calculated heat source intensity is too large, and the electrolyte resistance parameter in the Joule heat calculation needs to be appropriately reduced (reducing resistance reduces Joule heat), or the reaction heat parameter corresponding to the chemical field reaction rate needs to be reduced; if the simulated temperature is lower than the experimental temperature, then the opposite applies. If the bubble coverage deviation exceeds ±10%, adjust the activation energy parameter of the fluid field: if the simulated coverage is higher than the experimental coverage, it indicates that the bubble generation rate is too fast, and the activation energy needs to be appropriately increased (achieved by increasing the temperature correction coefficient); if the simulated coverage is lower than the experimental coverage, then the opposite applies. After parameter adjustment, recalculate the thermal and fluid fields to obtain new temperature and bubble coverage distributions, completing one thermal-fluid coupling iteration. The thermal-fluid iteration needs to be performed twice consecutively to ensure that the deviation of the core field gradually decreases.

[0036] The second step is the coordinated iteration of other fields: after every two heat flux iterations, the coordinated iteration of the electric field, chemical field, and structural field is initiated. In the electric field calibration, the simulated current density distribution data is compared with the experimental data collected by the sensor array. If the deviation exceeds 5%, the conductivity parameters of the electrolyte are adjusted (conductivity fluctuations may occur due to temperature changes), or the transfer coefficient of the chemical field is corrected (the transfer coefficient affects the correlation between reaction rate and current density). In the chemical field calibration, the reaction rate is recalculated based on the adjusted current density and fed back to the thermal and fluid fields to update the heat source term and bubble generation rate. In the structural field calibration, the simulated stress distribution data is compared with the experimental data collected by the stress sensor. If the deviation exceeds 10%, the material parameters of the structural field (such as the coefficient of thermal expansion and elastic modulus) are adjusted, and the electrode deformation is recalculated. If the deformation displacement exceeds 0.1 mm, it is fed back to the electric field model to adjust the electrode spacing, completing one coordinated iteration of other fields.

[0037] The logical judgments during the iteration process must ensure the flexibility and reliability of the calibration: First, dynamic adjustment of the convergence threshold—during the iteration process, the temperature gradient of the thermal field is calculated in real time (the temperature difference between two adjacent calculation units divided by the spatial distance). If the temperature gradient in a certain region exceeds 5℃ / cm (such as the electrode edge, where high temperature gradients are prone to occur due to current concentration), the temperature deviation threshold for that region is automatically tightened to ±1.5℃. After the temperature gradient in that region drops below 5℃ / cm, the initial threshold is restored to ensure the simulation accuracy of the high temperature gradient region; Second, handling of abnormal data—if the experimental data is abnormal at a certain moment (such as the current density exceeding the normal range of 500-2000A / m², or a sudden change in temperature data), the system automatically triggers an anomaly judgment mechanism: first, it checks the rationality of the data (such as whether it exceeds the measurement range of the equipment), and if so, it judges... If the data is identified as abnormal (e.g., due to sensor failure), the average data from the previous 5 time points is used to replace the data at that time point, and an audible and visual alarm signal is issued to remind the experimenters to check the equipment. If abnormal data occurs for 3 consecutive time points, the iteration is paused and resumed from the breakpoint of the last normal iteration after the equipment is repaired. The third is the judgment of iteration termination - after each iteration, the iteration is terminated if the following five conditions are met simultaneously: ① The temperature deviation of all key detection points in the thermal field (including 5 points on the electrode surface, 3 points in the electrolyte center, and 2 points on the tank wall) does not exceed ±2.5℃, and the temperature gradient region does not exceed ±1.5℃; ② The bubble coverage deviation of the fluid field does not exceed ±10%; ③ The electric field current density deviation does not exceed 5%; ④ The structural field stress deviation does not exceed 10%; ⑤ The above deviations remain stable for 3 consecutive iterations (the deviation change is less than 1%). If the condition is still not met after 50 iterations, analyze the physical field with the largest deviation (e.g., the temperature deviation of the thermal field is consistently large), check the governing equation parameters of that field (e.g., whether a certain type of heat is missing in the calculation of the heat source term) or the experimental equipment (e.g., whether the infrared thermal imager needs to be recalibrated), adjust, and restart the iteration; here, the deviation change refers to the relative rate of change of the deviation between two adjacent iterations, and the calculation formula is: Where δ is the change in deviation. This represents the deviation during the (n+1)th iteration. This represents the deviation during the nth iteration.

[0038] S3. Repeat step S2 until the deviation between the multi-field simulation results and the experimental data meets the preset convergence conditions. The preset convergence conditions include a thermal field temperature deviation of less than ±2.5℃ and a fluid field bubble coverage deviation of less than ±10%. The verified multi-field coupled simulation model is obtained. The preset convergence conditions are dynamically adjusted according to the simulation accuracy requirements. When the thermal field temperature gradient is greater than 5℃ / cm, the temperature deviation threshold is tightened to ±1.5℃.

[0039] This detailed embodiment elaborates on the internal operating logic of each major functional module, aiming to provide a detailed basis and explanation for those skilled in the art to understand and implement it. It should be emphasized that the above description constitutes a specific, preferred embodiment, but the concept of the present invention is not limited thereto. Any equivalent transformations, modifications, or improvements based on the core spirit of the present invention, without departing from the technical principles and scope disclosed in this specification, should be considered to fall within the scope of protection claimed by the present invention, as long as they achieve the same or similar technical effects.

Claims

1. An experimental verification method for multi-field intelligent simulation technology of a water electrolyzer, characterized in that, Includes the following steps: S1. Construct a multiphysics simulation model that includes a thermal field, a fluid field, and at least one other physical field selected from electric, chemical, or structural fields. The model achieves dynamic coupling through inter-field correlation parameters, wherein the coupling between the thermal field and the fluid field includes: The bubble generation kinetic parameters of the fluid field are dynamically corrected based on the temperature distribution of the thermal field. This correction uses the Arrhenius equation to correlate temperature with the activation energy of bubble generation. The activation energy Ea in the equation is expressed as... , where Ea0 is the reference activation energy and k is the temperature correction factor; The heat transfer coefficient of the thermal field is dynamically corrected based on the characteristic parameters of the bubble in the fluid field. These bubble characteristic parameters include bubble coverage, rising velocity, and morphology coefficient. The heat transfer coefficient h(φ) is expressed as... , where φ is the bubble coverage, h0 is the baseline heat transfer coefficient without bubbles, and α is the coverage correction factor; S2. The inter-field correlation parameters are corrected through a bidirectional iterative calibration module. The iterative calibration takes the coupling of the thermal field and the fluid field as the core and coordinates the correction of the correlation parameters of other physical fields. The bidirectional iterative calibration module prioritizes the coupling iteration of the thermal field and the fluid field, and the iteration frequency is not less than twice the coupling iteration frequency of other physical fields. S3. Repeat step S2 until the deviation between the multi-field simulation results and the experimental data meets the preset convergence conditions. The preset convergence conditions include a thermal field temperature deviation of less than ±2.5℃ and a fluid field bubble coverage deviation of less than ±10%, thus obtaining the verified multi-field coupled simulation model.

2. The experimental verification method for multi-field intelligent simulation technology of a water electrolyzer according to claim 1, characterized in that, The reference activation energy Ea0 mentioned in S1 ranges from 40 to 50 kJ / mol, and the temperature correction coefficient k ranges from 0.008 to 0.012 kJ / (mol·℃). Ea0 and k are obtained by fitting experimental data of bubble generation, and the experimental temperature range covers 20-80℃.

3. The experimental verification method for multi-field intelligent simulation technology of water electrolyzer according to claim 1, characterized in that, The range of values ​​for the reference heat transfer coefficient h0 mentioned in S1 is as follows: The coverage correction coefficient α ranges from 0.3 to 0.

5. h0 and α are determined by heat flux density measurement experiments. The experiments use a miniature heat flux sensor array with a sensor spacing of no more than 2 mm and a response time of no more than 5 ms.

4. The experimental verification method for multi-field intelligent simulation technology of a water electrolyzer according to claim 1, characterized in that, The bubble characteristic parameters described in S1 are acquired synchronously by a high-speed photography device and an infrared thermal imager. The frame rate of the high-speed photography device is not less than 1000fps, the temperature measurement accuracy of the infrared thermal imager is not less than ±1℃, and the spatial resolution is not less than 640×512 pixels.

5. The experimental verification method for multi-field intelligent simulation technology of a water electrolyzer according to claim 1, characterized in that, The correction of the associated parameters of other physical fields mentioned in S2 includes: correcting the chemical field reaction rate based on the electric field current density distribution, wherein the reaction rate is calculated by the Butler-Volmer equation, and the overpotential η of the equation is determined based on the current density distribution i(x,y) and the electrode material characteristic parameters. The correction of the associated parameters of other physical fields mentioned in S2 also includes: correcting the electric field electrode spacing based on the stress distribution of the structural field. This correction uses finite element analysis to convert the stress distribution σ(x,y) of the structural field into electrode deformation displacement Δd, thereby adjusting the electric field strength. Where d0 is the initial electrode spacing and U is the slot voltage.

6. The experimental verification method for multi-field intelligent simulation technology of a water electrolyzer according to claim 1, characterized in that, The bidirectional iterative calibration module described in S2 enables real-time data interaction between the thermal field simulation software and the fluid field simulation software via the Socket protocol. The data update delay is no more than 50ms, and the data format includes timestamps, spatial coordinates, and corresponding physical field parameters. It supports breakpoint resume and data verification.

7. The experimental verification method for multi-field intelligent simulation technology of a water electrolyzer according to claim 1, characterized in that, The preset convergence condition described in S3 is dynamically adjusted according to the simulation accuracy requirements. When the thermal field temperature gradient is greater than 5℃ / cm, the temperature deviation threshold is tightened to ±1.5℃.

8. The experimental verification method for multi-field intelligent simulation technology of water electrolyzer according to claim 2, characterized in that, The data acquisition conditions for the bubble generation experiment covered the current density. For each set of working conditions, at least three repeated experimental data should be collected, and the relative standard deviation of the experimental data should not exceed 5%.

9. The experimental verification method for multi-field intelligent simulation technology of a water electrolyzer according to claim 5, characterized in that, The overpotential η in the Butler-Volmer equation is determined based on the current density distribution i(x,y) and the electrode material characteristic parameters, including the exchange current density and the transfer coefficient.

10. The experimental verification method for multi-field intelligent simulation technology of a water electrolyzer according to claim 5, characterized in that, The coupling sequence between fields in the multiphysics simulation model is determined based on sensitivity analysis. The coupling iteration between the thermal field and the fluid field is completed twice, while the coupling iteration between other physical fields is completed once.

Images