A method for optimizing the design of the flow channel of an electrolytic cell assisted by machine learning technology

Abstract

The present invention relates to a method for optimizing the design of an electrolytic cell flow channel assisted by machine learning technology, belonging to the technical field of electrolytic water hydrogen production. The method comprises the following steps: S1. Based on the influence of gas-liquid mixed flow in an alkaline electrolytic water hydrogen production electrolytic cell on the electrochemical performance, a flow channel flow field model is selected to construct a basic model for optimizing the simulation of the flow channel; S2. A regression method is used to screen out the key variables of the flow channel geometric parameters; S3. The key variables are utilized and the flow field model is optimized through a genetic algorithm to obtain the parameters of the optimized key variables. This method takes the gas-liquid two-phase flow behavior inside the electrolytic cell as the key index of the electrolytic cell flow channel performance, optimizes the finally obtained flow channel design scheme, iteratively updates the combination of key variables based on the flow uniformity coefficient, reduces the flow dead zone or the excessive flow velocity in the local area, and promotes the discharge efficiency of hydrogen and oxygen.

Description

Technical Field

[0001] The present invention belongs to the technical field of hydrogen production by electrolyzing water, and particularly relates to a method for optimizing the design of an electrolyzer flow channel assisted by machine learning technology. Background Art

[0002] Hydrogen energy is an important part of the future energy system. Hydrogen has the advantage of high energy density, can effectively store a large amount of energy, and is suitable for application in large-scale energy storage and transportation. Moreover, hydrogen does not produce harmful pollutants during use and is a completely clean energy source. Producing hydrogen by electrolyzing water is a green hydrogen production method with zero carbon emissions. Different hydrogen production technologies by electrolyzing water have their own advantages and disadvantages, and alkaline water electrolysis has the advantages of mature technology and relatively low cost, and is one of the hydrogen production technologies with the fastest commercial progress currently.

[0003] An alkaline electrolyzer is the core equipment of the alkaline water electrolysis hydrogen production technology, which directly affects the hydrogen production efficiency and cost. In the research of alkaline water electrolysis hydrogen production, the design of the flow channel structure on the electrode plate is one of the key factors affecting the performance of the alkaline electrolyzer. The optimization goal of the flow channel design is mainly to improve the flow uniformity of the electrolyte and promote the smooth discharge of gas products, thereby effectively reducing the voltage loss of the electrolyzer.

[0004] At present, there are still some problems to be improved in the flow channel design of alkaline electrolyzers. First, there are flow dead zones or excessively high flow velocities in local areas in the flow channel, resulting in uneven distribution of the electrolyte, which directly affects the discharge efficiency of hydrogen and oxygen. Second, the current flow channel design mainly focuses on the analysis of the flow characteristics of the electrolyte, and optimizes the flow channel structure of the electrolyzer empirically based on the flow characteristics. The existing research on the optimization design of the alkaline electrolyzer flow channel is still in the initial stage, lacking a systematic optimization design method, and there is an urgent need to develop an efficient electrolyzer flow channel design optimization method to improve the performance of the electrolyzer and reduce the operating cost of electrolytic hydrogen production. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for optimizing the design of an electrolyzer flow channel assisted by machine learning technology to solve the problems such as flow dead zones or excessively high flow velocities in local areas existing in the existing alkaline electrolyzer flow channel design and the lack of a systematic optimization design method.

[0006] To achieve the above purpose, the technical solution of the present invention is as follows:

[0007] The present invention relates to a method for optimizing the design of an electrolyzer flow channel assisted by machine learning technology, which includes the following steps:

[0008] S1. Based on the influence of the gas-liquid two-phase flow in the alkaline water electrolysis hydrogen production electrolyzer on the electrochemical performance, construct a flow field model of the flow channel as the basic model for flow channel optimization simulation;

[0009] S2. Use the regression method to screen the key variables of the runner geometric parameters;

[0010] S3. Utilize the key variables and optimize the flow field model of the runner through the genetic algorithm to obtain the combination of optimized key variables, and then screen and optimize the runner structure.

[0011] Preferably, the specific steps for constructing the flow field model in S1 include:

[0012] S1.1. Construct the mass conservation formula of the fluid, and the mass conservation formula is expressed as:

[0013] (1),

[0014] where, φ g is the volume fraction of the gas phase, u g is the flow velocity of the gas phase, u l is the flow velocity of the liquid phase, m dc is the mass flux from the liquid phase to the gas phase, ρ l is the liquid phase density, ρ g is the gas phase density;

[0015] S1.2. Use the Navier-Stokes equation to describe the momentum change of the fluid in the electrolyzer runner, and the formula is:

[0016] (2),

[0017] (3),

[0018] where, ∇ p is the fluid pressure gradient, τ l and τ g are the stress tensors of the liquid phase and the gas phase respectively, F m，l is the momentum transferred from the liquid phase to the gas phase, F m，g is the momentum transferred from the gas phase to the liquid phase, u int is the fluid flow velocity inside the microelement;

[0019] S1.3. Use the k-ε equation to describe the gas-liquid two-phase turbulent flow behavior, and the formula is:

[0020] (4),

[0021] (5),

[0022] Among them, ε is the turbulent dissipation ratio, k is the turbulent kinetic energy, u mix is the fluid velocity, ρ is the fluid density, T is the ambient temperature of the reaction, μ T is the dynamic viscosity of the fluid, P k is the turbulent kinetic energy source term, σ k 、σ ε 、C ε,1 、C ε,2 are respectively the parameters of the k- ε turbulence model;

[0023] In formulas (4) and (5),

[0024] (6),

[0025] (7),

[0026] Among them, C u is also the parameter of the k- ε turbulence model.

[0027] Preferably, in the flow field model constructed in S1, on the anode side and the cathode side of the electrolytic cell flow channel, the mass flux from the liquid phase to the gas phase follows the following formula:

[0028] (8),

[0029] Among them, M g is the molar mass of the gas, n is the amount of electric charge transferred in the reaction, F is the Faraday constant, i ν is the reaction current of the microelement. The reaction current of the electrochemical reaction calibrates the gas mass flux of the microelement m g .

[0030] Preferably, the specific steps of using the regression method in S2 to screen the key variables of the flow channel geometric parameters are as follows:

[0031] S2.1. Establish an initial design of the electrolytic cell flow channel, and select several bump pitch parameters of the electrolytic cell as initial variables according to the behavior differences of the flow field in the flow channel;

[0032] S2.2. Establish several electrolytic cell flow channel bumps with different shapes, and select several initial variables of the flow channel shape geometric parameters according to the differences in the influence of bumps with different shapes on the flow field;

[0033] S2.3. Use multiple regression models to score each initial variable respectively, and normalize the scoring data to obtain the final score for each initial variable;

[0034] S2.4. Set a scoring threshold, and select the variables with the final score greater than the scoring threshold as key variables.

[0035] Preferably, in the S2.3, three regression models, namely regression tree, gradient boosting regression, and extra tree regression, are used to score each initial variable respectively. Among them, the flow channel result characteristics for constructing the regression method are called data dimensions, and the following formula is used for normalization processing to calculate the final score: (9),

[0036] where, X norm is the normalized data, X is the original data of the regression model scoring result, X min is the minimum value in each dimension of the original data, X max is the maximum value in each dimension of the original data.

[0037] Preferably, the specific steps of using the key variables in S3 to optimize the flow channel flow field model through the genetic algorithm include:

[0038] S3.1. Randomly combine the selected key variables to form several key variable combinations. The key variable combinations are used as individuals to generate the initial population of the genetic algorithm, and set the total number of iterations of the genetic algorithm;

[0039] S3.2. Conduct a flow field simulation of the alkaline electrolyzed water hydrogen production electrolytic cell for the flow channel structure corresponding to each individual, calculate the corresponding flow uniformity coefficient as the fitness of the individual, and calculate the total fitness value;

[0040] S3.3. Set the threshold of the number of individuals in the next generation population. Use the quotient of the individual fitness and the total fitness value as the probability of random selection. Randomly select a pair of individuals, and select the individual with higher fitness among the two and put it into the next generation population. Repeat this step until the number of individuals in the next generation population reaches the threshold of the number of individuals in the next generation population, and the number of iterations is incremented by 1;

[0041] S3.4. Determine whether the number of iterations reaches the set total number of iterations. If not, generate new individuals through the crossover method and / or mutation method and add them to the population, and return to S3.2 for the next round of iteration; if so, stop the iteration, output the optimized individuals, and thus obtain the optimized combination of key variables.

[0042] Preferably, the calculation formula for the flow uniformity coefficient in S3.2 is:

[0043] (10),

[0044] where, U v is the flow uniformity coefficient of the electrolyte, u i is the flow velocity of the i th flow channel microelement, u a is the weighted average flow velocity of the overall flow field, A i is the area of the i th flow channel microelement, n is the total number of flow channel microelements;

[0045] The calculation method of the weighted average flow velocity of the overall flow field is:

[0046] (11).

[0047] Adopting the technical solution provided by the present invention, compared with the prior art, it has the following beneficial effects:

[0048] 1. The method for optimizing the flow channel of an electrolytic cell assisted by machine learning technology involved in the present invention optimizes the flow channel morphology of the electrolytic cell under the flow model through a genetic algorithm based on the fluid flow field model in the alkaline electrolytic water hydrogen production electrolytic cell, obtains the optimized combination of key variables, and obtains the optimized design scheme, so that the finally obtained flow channel design scheme is optimized.

[0049] 2. The method for optimizing the flow channel of an electrolytic cell assisted by machine learning technology involved in the present invention iteratively updates the combination of key variables based on the flow uniformity coefficient during the process of optimizing the flow field model of the flow channel through a genetic algorithm, and thus obtains the optimal combination of key variables, thereby reducing the flow dead zone or the excessive flow velocity in the local area and promoting the discharge efficiency of hydrogen and oxygen. Description of the Drawings

[0050] Figure 1 is the schematic flow chart of the method of the present invention;

[0051] Figure 2 Schematic diagram of the flow channel partition method of the electrolytic cell in the embodiment;

[0052] Figure 3 Schematic diagram of the geometric dimension parameter names of the flow channel turbulence unit distribution in the embodiment;

[0053] Figure 4 Importance score diagram of the regression method for the geometric parameters of the turbulence unit geometry structure after data normalization in the embodiment;

[0054] Figure 5 Schematic diagram of the genetic algorithm optimization process for the flow channel structure design in the embodiment;

[0055] Figure 6 Data diagram of the genetic algorithm optimization process for the flow channel optimization design of the electrolytic cell assisted by machine learning technology in the embodiment;

[0056] Figure 7 Diagram of the best flow channel structure and flow field flow velocity distribution for the flow channel optimization design of the electrolytic cell assisted by machine learning technology in the embodiment.

[0057] Figure 8 Data diagram of the optimized flow uniformity coefficient for the flow channel optimization design of the electrolytic cell assisted by machine learning technology in the embodiment. Detailed implementation manners

[0058] To further understand the content of the present invention, the present invention will be described in detail in combination with the embodiments. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0059] Refer to the attached Figure 1 As shown, the present invention relates to a method for optimizing the design of an electrolytic cell flow channel assisted by machine learning technology, which includes the following steps:

[0060] S1. Based on the influence of the gas-liquid mixed flow in the alkaline electrolytic water hydrogen production electrolytic cell on the electrochemical performance, by partially simplifying the non-critical factors of the model and focusing on the core contradiction of the optimization process, a fluid flow field model in the flow channel is constructed as the basic model for flow channel optimization simulation;

[0061] Among them, the non-critical factors include the following assumptions:

[0062] (1) The water inside the electrolytic cell always exists in a liquid state, and the evaporation phenomenon of the liquid water is ignored;

[0063] (2) All gases in the electrolytic cell are ideal gases and are incompressible;

[0064] (3) There is no significant contact resistance between the layers of the electrolytic cell;

[0065] (4) The cross-permeation phenomenon of hydrogen and oxygen in the diaphragm can be ignored.

[0066] The above-mentioned construction of the flow field model uses the Eulerian-Eulerian Reynolds-Averaged Navier-Stokes k-ε turbulence model to calculate the fluid velocity and gas volume ratio in the flow channel of the electrolytic cell. By adding a viscosity term correction in the large-scale motion to reflect the influence of small-scale motion, the simulation accuracy of the model in rotational flow is improved, and at the same time, the simulation accuracy of the flow conditions in the turbulent, laminar, and laminar-turbulent transition regions in the flow field model is improved. The specific steps of the construction method of the flow field model include:

[0067] S1.1. Construct the mass conservation formula of the fluid, and the mass conservation formula is expressed as:

[0068] (1),

[0069] Among them, φ g is the volume fraction of the gas phase, u g is the velocity of the gas phase, u l is the velocity of the liquid phase, m dc is the mass flux from the liquid phase to the gas phase, ρ l is the liquid phase density, ρ g is the gas phase density;

[0070] S1.2. To describe the flow behavior of the fluid, the Navier-Stokes equation is used to describe the momentum change of the fluid in the flow channel of the electrolytic cell, and the formula is:

[0071] (2),

[0072] (3),

[0073] Among them, ∇ p is the fluid pressure gradient, τ l and τ g are the stress tensors of the liquid phase and the gas phase respectively, F m，l is the momentum transferred from the liquid phase to the gas phase, F m，g is the momentum transferred from the gas phase to the liquid phase, u int is the fluid velocity inside the microelement;

[0074] S1.3. Use the k-ε equation to describe the turbulent flow behavior of the gas-liquid two-phase, and the formula is:

[0075] (4),

[0076] (5),

[0077] where, ε is the turbulent dissipation ratio, k is the turbulent kinetic energy, u mix is the fluid velocity, ρ is the fluid density, T is the ambient temperature of the reaction, μ T is the dynamic viscosity of the fluid, P k is the turbulent kinetic energy source term, σ k 、σ ε 、C ε,1 、C ε,2 are respectively the parameters of the k - ε turbulence model;

[0078] In formulas (4) and (5),

[0079] (6),

[0080] (7),

[0081] where, C u is also the parameter of the k - ε turbulence model;

[0082] As a supplement, on the anode side and the cathode side of the electrolytic cell flow channel, the mass flux from the liquid phase to the gas phase follows the following formula:

[0083] (8),

[0084] where, M g is the molar mass of the gas, n is the amount of electric charge transferred in the reaction, F is the Faraday constant, i ν is the reaction current of the micro - element. The reaction current of the electrochemical reaction calibrates the gas mass flux of the micro - element m g .

[0085] S2. Use the regression method to screen the key variables of the flow channel geometric parameters. The specific steps are as follows:

[0086] S2.1. Establish an initial electrolytic cell flow channel design, and partition it according to the differences in the behavior of the flow field within the flow channel. The partitioning results are as shown in Figure 2 ; classify and label the geometric structures of the flow channel bypass units within the partition, as shown in Figure 3 ; select the bump pitch parameters of 16 electrolytic cells as the initial variables of the flow channel structure parameters;

[0087] S2.2. Establish several electrolytic cell flow channel bumps with different shapes, and select several initial variables of the geometric parameters of the flow channel shape according to the differences in the influence of bumps with different shapes on the flow field;

[0088] S2.3. Use multiple regression models to score each initial variable (including the initial variables of the flow channel structure parameters and the initial variables of the geometric parameters of the flow channel shape). In this embodiment, three regression methods, namely regression tree, gradient boosting regression, and extra tree regression, are used to screen the key variables. After numerically normalizing the evaluation results of the three regression methods, a comprehensive evaluation is obtained, which can avoid the deviation that may be brought by a single regression model. The above three regression methods are all conventional technical means in the art and do not belong to the content protected by this application. Normalize the scoring data, calculate and obtain the final score for each initial variable as the importance score. Among them, the flow channel result characteristics for constructing the regression method are called data dimensions, and the calculation formula is as follows:

[0089] (20),

[0090] where, X norm is the normalized data, X is the original data of the regression model scoring result, X min is the minimum value in each dimension of the original data, X max is the maximum value in each dimension of the original data.

[0091] S2.4. Set a scoring threshold, and select the variables with the final score greater than the scoring threshold as the key variables.

[0092] After S2, obtain the importance scores of the flow channel structure parameters as shown in Figure 4 . Select the flow channel structure parameters exceeding the scoring threshold as the key variables. In the embodiment, 6 flow channel structure parameters are selected as the key variables; similarly, several geometric parameters of the flow channel shape can be selected as the key variables, and the sum of the two types of key variables is all the selected key variables.

[0093] S3. Use the key variables and optimize the flow channel flow field model through the genetic algorithm to obtain a combination of optimized key variables, and then screen and optimize the flow channel structure. The specific steps are as shown in Figure 5 and include:

[0094] S3.1. Randomly combine the selected key variables to form several key variable combinations. The key variable combinations serve as the initial population of the genetic algorithm for individuals, and set the total number of iterations of the genetic algorithm.

[0095] S3.2. Conduct a flow field simulation of the alkaline electrolyzed water hydrogen production electrolyzer for the flow channel structure corresponding to each individual, calculate the corresponding flow uniformity coefficient as the fitness of the individual, and calculate the total fitness value. The calculation formula for the flow uniformity coefficient of each individual is:

[0096] (21),

[0097] Where, U v is the flow uniformity coefficient of the electrolyte, u i is the flow velocity of the i th flow channel microelement, u a is the weighted average flow velocity of the overall flow field, A i is the i th area of the flow channel microelement, n is the total number of flow channel microelements;

[0098] The calculation method of the weighted average flow velocity of the overall flow field is:

[0099] (22).

[0100] S3.3. Set the threshold for the number of individuals in the next generation population. Use the quotient of the individual fitness and the total fitness value as the probability of random selection. Randomly select a pair of individuals, and select the individual with higher fitness among the two and put it into the next generation population. Repeat this step until the number of individuals in the next generation population reaches the threshold for the number of individuals in the next generation population, and increase the iteration count by 1. Improve the retention chance of individuals with high fitness through the selection method;

[0101] S3.4. Determine whether the iteration count has reached the set total number of iterations. If not, generate new individuals through the crossover method and / or mutation method and add them to the population. Introduce new configuration combinations through the crossover and mutation methods to increase the population diversity. Specifically, the crossover method is to exchange the numerical values of the spacing parameters at the corresponding positions of two individuals, and the mutation method is to randomly select several spacing parameters in an individual with a small probability and randomly generate a spacing parameter to replace the original numerical value of the spacing parameter of the individual, and return to S3.2 for the next round of iteration; if so, stop the iteration and output the optimized individual, thereby obtaining the optimized combination of key variables. In this embodiment, the optimization result is as Figure 6 shown.

[0102] Effect analysis: After being optimized by the genetic algorithm, the optimization process converges after the 38th generation, the flow uniformity coefficient reaches the minimum value and tends to be stable, indicating that the flow channel structure has the optimal flow uniformity at this time. The geometric structure of the flow channel and the electrolyte flow velocity distribution are as follows Figure 7 shown. The comparison of the flow uniformity coefficients of various flow channels before optimization at four characteristic flow channel cross-sections is as follows Figure 8 shown. The flow uniformity coefficient of the optimal flow channel structure is 0.6920, which is 1.97% higher than that of Model_3.4, 9.09% and 13.80% higher than those of DFC and CCSFC respectively, and this flow uniformity coefficient is significantly higher than the results of the single-factor analysis of all unoptimized initial flow channel parameters.

[0103] The present invention has been described in detail above in conjunction with the embodiments, but the above content is only the preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. Any equivalent changes and improvements made within the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. A method for optimizing the flow channel design of an electrolytic cell based on machine learning technology, characterized in that: It includes the following steps: S1. Based on the influence of gas-liquid mixed flow in the alkaline water electrolysis hydrogen production electrolyzer on the electrochemical performance, a flow field model of the flow channel is constructed as the basic model for flow channel optimization simulation; S2. Use regression method to screen key variables of flow channel geometry parameters. The specific steps are as follows: S2.

1. Establish an initial electrolytic cell flow channel design, and select several electrolytic cell convex point spacing parameters as initial variables of the flow channel structure parameters according to the differences in the behavior of the flow field in the flow channel; S2.

2. Establish several electrolytic cell flow channel convex points of different shapes, and select several initial variables of flow channel shape geometric parameters according to the difference in the influence of different shape convex points on the flow field; S2.

3. Use multiple regression models to score each initial variable respectively, and normalize the scoring data to obtain the final score for each initial variable; S2.

4. Set a scoring threshold and select variables whose final scores are greater than the scoring threshold as key variables; S3. Utilize key variables and optimize the flow field model of the flow channel through genetic algorithm to obtain the optimized combination of key variables, and then screen and optimize the flow channel structure. The specific steps include: S3.

1. Randomly combine the selected key variables to form several key variable combinations, and the key variable combinations are used as the initial population of the genetic algorithm to generate individuals, and the total number of iterations of the genetic algorithm is set; S3.

2. Perform alkaline water electrolysis hydrogen production electrolyzer flow field simulation on the flow channel structure corresponding to each individual, calculate the corresponding flow uniformity coefficient as the fitness of the individual, and calculate the total fitness value; S3.

3. Set the threshold of the next generation population size, take the quotient of the individual fitness and the total fitness value as the probability of random selection, randomly select a pair of individuals, and select the individual with higher fitness between the two and put it into the next generation population, repeat this step until the number of individuals in the next generation population reaches the threshold of the next generation population size, and the number of iterations increases by 1; S3.

4. Determine whether the number of iterations reaches the set total number of iterations. If not, generate new individuals by crossover method and / or mutation method and add them to the population, and return to S3.2 for the next round of iterations; If it is reached, stop the iteration, output the optimized individual, and then obtain the optimized combination of key variables.

2. The method for optimizing the flow channel of an electrolytic cell based on machine learning technology according to claim 1, characterized in that: The specific steps of constructing the flow field model in S1 include: S1.

1. Construct the mass conservation formula of the fluid. The mass conservation formula is expressed as: （1）， in, φ g is the volume fraction of the gas phase, u g is the flow rate of the gas phase, u l is the flow rate of the liquid phase, m dc is the mass flux from liquid to gas, ρ l is the liquid density, ρ g is the gas phase density; S1.

2. The Navier-Stokes equation is used to describe the momentum change of the fluid in the electrolytic cell flow channel. The formula is: （2）， （3）， Among them, ∇ p is the fluid pressure gradient, τ l and τ g are the stress tensors of the liquid and gas phases, respectively, F m，l is the momentum transferred from the liquid phase to the gas phase, F m，g is the momentum transferred from the gas phase to the liquid phase, u int is the flow velocity of the fluid inside the microelement; S1.

3. The k-ε equation is used to describe the gas-liquid two-phase turbulent flow behavior, and the formula is: （4）， （5）， in, ε is the turbulence dissipation ratio, k is the turbulent kinetic energy, u mix is the fluid flow rate, ρ is the fluid density, T is the ambient temperature of the reaction, μ T is the fluid dynamic viscosity, P k is the turbulent flow energy term, σ k , σ ε 、C ε,1 、C ε,2 They are k- ε Parameters of the turbulence model; In formula (4) and formula (5), （6）， （7）， in, C u Also k- ε Parameters of the turbulence model.

3. The method for optimizing the flow channel of an electrolytic cell based on machine learning technology according to claim 2, characterized in that: In the flow field model constructed in S1, the mass flux from the liquid phase to the gas phase on the anode side and the cathode side of the electrolytic cell flow channel follows the following formula: （8）， in, M g is the molar mass of the gas, n is the amount of charge transferred by the reaction, F is the Faraday constant, i ν is the reaction current of the microelement, and the gas mass flux of the microelement is calibrated by the reaction current of the electrochemical reaction m g .

4. The method for optimizing the flow channel of an electrolytic cell based on machine learning technology according to claim 1, characterized in that: In S2.3, three regression models, namely regression tree, gradient boosting regression and extra tree regression, are used to score each initial variable respectively, wherein the flow channel result feature of the regression method is called data dimension, and the normalization process uses the following formula to calculate the final score: (9), in, X norm To normalize the data, X The raw data for the regression model scoring results, X min is the minimum value of each dimension in the original data. X max is the maximum value of each dimension in the original data.

5. The method for optimizing the flow channel of an electrolytic cell based on machine learning technology according to claim 1, characterized in that: The calculation formula of the flow uniformity coefficient in S3.2 is: （10）， in, U v is the flow uniformity coefficient of the electrolyte, u i For the i The flow velocity of a microelement of the flow channel is u a is the weighted mean of the velocity of the entire flow field, A i For the i The area of ​​a flow channel element, n is the total number of flow channel elements; The calculation method of the weighted mean of the flow velocity of the entire flow field is: （11）。

Images