Solid oxide fuel cell flow channel structure and multi-objective optimization design method thereof
By employing a flow channel structure with alternating split and merge units in a solid oxide fuel cell, and utilizing surrogate models and multi-objective genetic algorithms for optimization design, the problems of large flow field pressure drop and uneven distribution of reactant gases in the flow field structure were solved, achieving efficient mass transfer of reactant gases and improved battery performance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- YUSHI ENERGY NANTONG CO LTD
- Filing Date
- 2023-04-03
- Publication Date
- 2026-07-03
AI Technical Summary
While improving output power, existing solid oxide fuel cell flow field structures suffer from problems such as large flow pressure drop, excessive mechanical stress and parasitic power, and uneven distribution of reactant gases.
A flow channel structure with alternating flow splitting and merging units is adopted. The design is optimized using a surrogate model and a multi-objective genetic algorithm to improve the uniform distribution of reactant gases and heat dissipation performance.
It achieves uniform distribution and efficient mass transfer of reactant gases, reduces process pressure drop and mechanical stress, and improves battery output power and performance.
Smart Images
Figure CN116387550B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of solid oxide fuel cell technology, and in particular to a solid oxide fuel cell flow channel structure and its multi-objective optimization design method. Background Technology
[0002] Solid oxide fuel cells (SOFCs) are power generation devices that directly convert the chemical energy of reactants into electrical energy. Compared with other types of fuel cells, SOFCs have advantages such as reliable all-solid-state component structure, high energy conversion efficiency, flexible fuel use, and low manufacturing cost, and are widely considered to be the most promising fuel cell technology. Currently, solid oxide fuel cells still have significant development potential in terms of lifespan and battery performance. The supply capacity and uniformity of the reactant gas are key factors affecting their performance.
[0003] The flow field structure of SOFCs has a significant impact on the distribution of reactant gases and component transport within them. An efficient flow field structure promotes uniform distribution of reactants and heat dissipation, thereby improving battery performance. Traditional flow field structures, such as parallel and serpentine flow fields, are largely optimized through empirical modifications to their channel structures. Research on aspects like channel length, ridge width, and the width ratio of the channel to the ridge is quite extensive, but its drawbacks are also apparent. For example, serpentine flow fields typically exhibit a more uniform flow distribution on the surface of the membrane electrode assembly, resulting in higher battery output power. However, serpentine flow fields generate significant flow field pressure drops, leading to additional mechanical stress and substantial parasitic power. Parallel flow fields suffer from severe uneven flow distribution, causing uneven heat and current generation, which can degrade battery performance. Therefore, how to reduce flow field pressure drops, avoid additional mechanical stress and parasitic power, and promote uniform flow distribution while increasing output power has become a pressing issue. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this invention provides a solid oxide fuel cell flow channel structure and its multi-objective optimization design method, aiming to improve the uniformity of the battery flow field and heat dissipation, thereby enhancing battery performance.
[0005] The technical solution adopted in this invention is as follows:
[0006] This application provides a flow channel structure for a solid oxide fuel cell, including a flow splitting unit and a flow combining unit. The flow splitting unit and the flow combining unit are arranged alternately along the flow direction, that is, the outlet of the front flow splitting unit is connected to the inlet of the rear flow combining unit, and the outlet of the rear flow combining unit is connected to the inlet of the next rear flow splitting unit.
[0007] The diversion unit includes an inlet section and a contraction section arranged sequentially along the flow direction. The outlet of the inlet section is connected to the inlet of the contraction section. The height of the contraction section gradually decreases and the width gradually increases along the flow direction. The height of the converging unit gradually increases along the flow direction.
[0008] The further technical solution is as follows:
[0009] The inlet height of the combiner unit is equal to the outlet height of the splitter unit, and the outlet height of the combiner unit is equal to the inlet height of the splitter unit.
[0010] Each row of branch units consists of 6-10 units arranged in parallel, and each row of bus units consists of one unit.
[0011] The structural dimensions of each branch unit are consistent, and the structural dimensions of each bus unit are consistent.
[0012] The structural parameters of the flow channel structure are obtained using a multi-objective optimization method based on a surrogate model. The optimization objectives of the multi-objective optimization are to optimize the net power density and oxygen uniformity of the solid oxide fuel cell. The surrogate model is obtained by training an artificial neural network and is used to establish the mapping relationship between the structural parameters and the objective function of the multi-objective optimization.
[0013] A second aspect of this application provides a solid oxide fuel cell, including an anode flow channel and a cathode flow channel, wherein the anode flow channel and / or the cathode flow channel adopts the flow channel structure of a solid oxide fuel cell as described above.
[0014] A third aspect of this application provides a multi-objective optimization design method for the flow channel structure of a solid oxide fuel cell as described above, comprising:
[0015] Based on the optimization objective, an objective function is established, and optimization variables and their corresponding value ranges are determined. The optimization objective is to optimize the net power density and oxygen uniformity of the solid oxide fuel cell, and the optimization variables are the structural parameters of the flow channel structure.
[0016] The Latin hypercube sampling method was used to select design points, and the objective function values of each design point were calculated by CFD numerical simulation to form a database.
[0017] The surrogate model is obtained by training an artificial neural network through the database. The input of the neural network is the optimization variable, and the output is the objective function, namely, the net power density of the battery and the oxygen uniformity.
[0018] Based on the aforementioned proxy model, a multi-objective optimization algorithm is used to obtain the structural parameters corresponding to the optimal battery net power density and oxygen uniformity.
[0019] The further technical solution is as follows:
[0020] Based on the aforementioned surrogate model, a multi-objective genetic algorithm is used to obtain the structural parameters corresponding to the optimal battery net power density and oxygen uniformity, including:
[0021] Initialize the population, the number of individuals in the population corresponds to the number of groups of the optimization variables, and each group of optimization variables is randomly selected within the range of the values stated;
[0022] The surrogate model is used to calculate the objective function corresponding to each group of optimization variables, and the objective function is used to evaluate the fitness of individuals within the population.
[0023] The offspring population is generated through selection, crossover, and mutation;
[0024] The parent and offspring populations are combined, and the next generation population is formed by non-dominated sorting and crowding calculation.
[0025] Repeat the above steps until the stopping criterion is met. At the end of the iteration, multiple Pareto nondominated solutions are obtained.
[0026] The optimal structural parameters are obtained by applying the order preference similarity technique of the ideal solution.
[0027] The optimization variables include: the length of the inlet section of the diversion unit, the length of the converging section of the diversion unit, the height of the outlet of the converging section of the diversion unit, the width of the outlet of the converging section of the diversion unit, and the length of the converging unit.
[0028] The total length of the bus unit and its adjacent branch unit is a set value.
[0029] The beneficial effects of this invention are as follows:
[0030] The flow channel structure of this application promotes continuous splitting and merging of the reactant gas, enhancing turbulence and facilitating a uniform distribution of the reactant gas flow field. This results in advantages such as high mass transfer efficiency, high output power, and rapid heat dissipation. The splitting unit increases the local velocity of the reactant gas; based on the field synergy principle, it transforms the transport of the reactant gas into the battery interior from diffusion to convection, thus improving mass transfer efficiency. The merging unit's cross-sectional area gradually increases along the reactant gas flow direction, effectively reducing vortex generation and improving the heat dissipation performance of the flow field.
[0031] Compared with traditional batteries using a serpentine flow field, the solid oxide fuel cell of this application has a smaller pressure drop, lower mechanical stress, higher output power, and significantly improved battery performance.
[0032] This application utilizes a surrogate model trained on a neural network and employs a multi-objective genetic algorithm to quantify and obtain the optimal structural parameters of the flow channel structure, thereby satisfying the objectives of optimal net power density and oxygen uniformity of the battery. The design method is reasonable, computationally efficient, and yields highly accurate results. The optimized fuel cell design can be widely applied in new energy equipment, including new energy ships, new energy aircraft, and new energy high-speed trains, meeting their application requirements.
[0033] Other features and advantages of the invention will be set forth in the following description or may be learned by practicing the invention. Attached Figure Description
[0034] Figure 1 This is a schematic diagram of the structure of Embodiment 1 of the present invention.
[0035] Figure 2 for Figure 1 Top view.
[0036] Figure 3 This is a schematic diagram of the structure of the shunt unit in Embodiment 1 of the present invention.
[0037] Figure 4 This is an exploded structural diagram of Embodiment 2 of the present invention.
[0038] Figure 5 This is a flowchart of the multi-objective optimization design method in Embodiment 3 of the present invention.
[0039] Figure 6 This is a structural diagram of the artificial neural network proxy model constructed in Embodiment 3 of the present invention.
[0040] Figure 7 This is the objective function response surface diagram generated using a surrogate model in Embodiment 3 of the present invention.
[0041] Figure 8 This is a flowchart of the genetic algorithm used in Embodiment 3 of the present invention.
[0042] Figure 9 This is the Pareto front and decision graph obtained using a genetic algorithm in Embodiment 3 of the present invention.
[0043] Figure 10 The image shows the performance curve of the solid oxide fuel cell prepared based on Example 3 of the present invention.
[0044] In the diagram: 1. Anode connector; 2. Anode flow channel; 3. Anode support layer; 4. Anode functional layer; 5. Electrolyte; 6. Cathode functional layer; 7. Cathode current collector layer; 8. Cathode flow channel; 9. Cathode connector; 10. Diverting unit; 11. Combining unit; 101. Inlet section; 102. Contraction section. Detailed Implementation
[0045] The specific embodiments of the present invention are described below with reference to the accompanying drawings.
[0046] Example 1
[0047] See Figure 1 and Figure 2 Embodiment 1 of this application provides a flow channel structure for a solid oxide fuel cell, including a flow splitting unit 10 and a flow combining unit 11. The flow splitting unit 10 and the flow combining unit 11 are arranged alternately in sequence along the flow direction, that is, the outlet of the front flow splitting unit 10 is connected to the inlet of the rear flow combining unit 11, and the outlet of the rear flow combining unit 11 is connected to the inlet of the next rear flow splitting unit 10; the height of the flow combining unit 11 gradually increases along the flow direction.
[0048] See Figure 3 The diversion unit includes an inlet section 101 and a constriction section 102 arranged sequentially along the flow direction. The inlet of the inlet section 101 is the inlet of the diversion unit, and the outlet of the constriction section 102 is the outlet of the diversion unit. The outlet of the inlet section 101 is connected to the inlet of the constriction section 102. The height of the constriction section 102 gradually decreases along the flow direction, and the width gradually increases along the flow direction.
[0049] In this application, "length" and "width" refer to the length and width of the object, respectively. Figure 2 The directions that are parallel or perpendicular to the flow direction.
[0050] It can be understood that in this embodiment, the inlet section of the diversion unit is rectangular, the contraction section is horn-shaped, and the converging unit is horn-shaped. See also Figure 1 In this embodiment, the inlet height of the combiner unit 11 is equal to the outlet height of the splitter unit, and the outlet height of the combiner unit is equal to the inlet height of the splitter unit.
[0051] Preferably, in this embodiment, the structural dimensions of each shunt unit are consistent, and the structural dimensions of each combiner unit are consistent.
[0052] In this embodiment, the solid oxide fuel cell flow channel structure adopts a "trapezoidal mesh" structure. Within this trapezoidal mesh flow field, splitting and merging units are arranged alternately, which promotes continuous splitting and merging of the reactant gas, enhancing turbulence, promoting uniform distribution of the reactant gas in the flow field, and facilitating mass transfer. Specifically, the splitting units increase the local velocity of the reactant gas; based on the field synergy principle, the transport of the reactant gas into the cell interior can be transformed from diffusion to convection, contributing to improved mass transfer efficiency. Meanwhile, the cross-sectional area of the merging units gradually increases along the flow direction of the reactant gas, effectively reducing vortex generation and improving the heat dissipation performance of the flow field. In summary, this trapezoidal mesh flow field has the advantages of high mass transfer efficiency, high output power, and rapid heat dissipation.
[0053] Specifically, each row of distribution units may include 6-10 units arranged in parallel, and each row of bus units includes one unit.
[0054] Example 2
[0055] Embodiment 2 of this application provides a solid oxide fuel cell, including an anode flow channel and a cathode flow channel, wherein the anode flow channel and / or cathode flow channel adopts the flow channel structure described in Embodiment 1 above.
[0056] See Figure 4 This embodiment of the solid oxide fuel cell includes an anode connector 1, an anode flow channel 2, an anode support layer 3, an anode functional layer 4, an electrolyte 5, a cathode functional layer 6, a cathode current collector 7, a cathode flow channel 8, and a cathode connector 9. The anode flow channel 2 is formed between the anode connector 1 and the anode support layer 3, and the cathode flow channel 8 is formed between the cathode current collector 7 and the cathode connector 9. The cathode flow channel 8 adopts the flow channel structure described in Embodiment 1, exhibiting a "trapezoidal mesh" structure, while the anode flow channel 2 can adopt a parallel flow channel structure. Utilizing the "trapezoidal mesh" structure, the solid oxide fuel cell of this embodiment can promote uniform distribution of reactants and uniform heat dissipation. Compared with traditional cells using a serpentine flow field, it has a smaller pressure drop, lower mechanical stress, higher output power, and significantly improved battery performance.
[0057] Example 3
[0058] To quantitatively design the flow channel structure and obtain the optimal flow channel structure, Embodiment 3 of this application provides a multi-objective optimization design method for the flow channel structure of the solid oxide fuel cell, see [link to documentation]. Figure 5 This includes the following steps:
[0059] S1. Establish an objective function and determine the optimization variables and their corresponding value ranges based on the optimization objective. The optimization objective is to optimize the net power density and oxygen uniformity of the solid oxide fuel cell. The optimization variables are the structural parameters of the flow channel structure.
[0060] S2. The Latin hypercube sampling method is used to select design points, and the objective function values of each design point are calculated by CFD (Computational Fluid Dynamics) numerical simulation to form a database.
[0061] S3. Train an artificial neural network through the database to obtain a surrogate model. The input of the neural network is the optimization variable, and the output is the objective function, namely the net power density of the battery and the oxygen uniformity, thereby establishing a mapping relationship between the structural parameters and the objective function of multi-objective optimization.
[0062] S4. Based on the surrogate model, a multi-objective optimization algorithm is used to obtain the structural parameters corresponding to the optimal battery net power density and oxygen uniformity.
[0063] Specifically, the objective function is the net power density P of the solid oxide fuel cell. net With oxygen uniformity The mathematical expression is as follows:
[0064] P net =P cell -P con (1)
[0065]
[0066] In equation (1), the battery output power P cell =U cell I ave Pressure loss power density P con =Δp cat u cat A ch,cat / A act U cell I is the battery output voltage. ave For the average current density, Δp cat u cat A ch,cat These represent the cathode voltage drop, inlet flow velocity, and inlet cross-sectional area of the fuel cell, respectively. act It is the active reaction area of the battery;
[0067] In equation (2), A i φ is the area of the i-th discrete region within the battery flow field. i It is the molar concentration of substance φ in the i-th discrete region. It is the average molar concentration of substance φ in each discrete region;
[0068] The optimization variables and their corresponding value ranges are as follows:
[0069] 0.3mm≤L is ≤0.8mm
[0070] 0.3mm≤L ss ≤0.8mm
[0071] 0.1mm≤D ss ≤0.7mm
[0072] 0.8mm≤W ss ≤1.7mm
[0073] L is L is the length of the inlet section of the diversion unit. ss D is the length of the contraction section of the diversion unit; ss W is the outlet height of the converging section of the diversion unit. ss The width of the outlet of the contraction section of the diversion unit.
[0074] For details, see Figure 6The artificial neural network is a single-layer feedforward backpropagation network, consisting of an input layer, a hidden layer, and an output layer. The number of neurons in the input layer is equal to the number of optimization variables. In this embodiment, the optimization variables include four: the length of the shunt unit inlet segment, the length of the shunt unit constriction segment, the height of the shunt unit constriction segment outlet, the width of the shunt unit constriction segment outlet, and the length of the shunt unit. The number of neurons in the output layer is 2, corresponding to the battery net power density and oxygen uniformity.
[0075] See Figure 7 The objective function response surface plot obtained using the surrogate model constructed in this embodiment is shown below. Figure 7 In the middle (a) and (b), the net power density P of the fuel cell is shown. net With oxygen uniformity The response surface diagram.
[0076] Specifically, this embodiment uses a multi-objective genetic algorithm as the multi-objective optimization algorithm to obtain the structural parameters corresponding to the optimal battery net power density and oxygen uniformity. See [link to relevant documentation]. Figure 8 The process includes the following steps:
[0077] Initialize the population, the number of individuals N in the population corresponds to the number of groups of the optimization variables, and each group of optimization variables is randomly selected within its value range;
[0078] The surrogate model is used to calculate the objective function corresponding to each group of optimization variables, and the objective function is used to evaluate the fitness of individuals within the population.
[0079] The offspring population is generated through selection, crossover, and mutation;
[0080] The parent and offspring populations are combined, and the next generation population is formed by non-dominated sorting and crowding calculation.
[0081] Repeat the above steps until the stopping criterion is met. At the end of the iteration, multiple Pareto nondominated solutions are obtained.
[0082] The optimal structure parameters are obtained by applying the order preference similarity technique of the ideal solution (i.e., the ordering method TOPSIS that approximates the ideal solution).
[0083] See Figure 9 The diagram shows the Pareto front and decision graph obtained by the genetic algorithm in this embodiment, with the optimal structural parameters corresponding to the best decision point in the graph. Specifically, the total length of the merging unit and its adjacent branching units is a set value, which is 2mm in this embodiment, i.e., the length L of the merging unit. col =2mm-(L is +L ss The optimal flow field structure parameters obtained are shown in Table 1.
[0084] Table 1 Results of Multi-Objective Optimization
[0085]
[0086] Further settings were made for other parameters: the inlet section height of each diversion unit was consistent at 0.8 mm; the inlet section width was consistent at 1.0 mm.
[0087] The solid oxide fuel cell of Example 2 was designed using the optimal structural parameters obtained in Example 3, and its performance was verified below.
[0088] The battery operates at one standard atmosphere and at a temperature of 1023 K. Humidified hydrogen gas is introduced through the anode at a relative humidity of 2% and a mass flow rate of 9.6 × 10⁻⁶. -8 kg / s; air is introduced into the cathode, with an inlet mass flow rate of 2.75 × 10⁻⁶ kg / s; -6 kg / s, the results were obtained through CFD simulation.
[0089] The performance curves of solid oxide fuel cells are as follows: Figure 10 As shown. Figure 10 Figures (a) and (b) show the battery polarization curve and battery output power curve, respectively. As can be seen from the figures, the current density of the solid oxide fuel cell increases as the voltage decreases. At a voltage of 0.4V, the maximum current density is 1.375 A / cm². 2 The net power density initially increases and then decreases with increasing current density, reaching a maximum output power of 0.5616 W / cm² at a voltage of 0.45V. 2 .
[0090] The solid oxide fuel cell designed using the optimized design method of this application can be widely used in new energy equipment, including new energy ships, new energy aircraft, and new energy high-speed trains.
[0091] It will be understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A flow channel structure for a solid oxide fuel cell, characterized in that, It includes a diversion unit and a confluence unit, which are arranged alternately in sequence along the flow direction. The outlet of the front diversion unit is connected to the inlet of the rear confluence unit, and the outlet of the rear confluence unit is connected to the inlet of the next rear diversion unit. The diversion unit includes an inlet section and a contraction section arranged sequentially along the flow direction. The outlet of the inlet section is connected to the inlet of the contraction section. The height of the contraction section gradually decreases and the width gradually increases along the flow direction. The height of the converging unit gradually increases along the flow direction.
2. The solid oxide fuel cell flow channel structure according to claim 1, characterized in that, The inlet height of the combiner unit is equal to the outlet height of the splitter unit, and the outlet height of the combiner unit is equal to the inlet height of the splitter unit.
3. The solid oxide fuel cell flow channel structure according to claim 1, characterized in that, Each row of branch units consists of 6-10 units arranged in parallel, and each row of bus units consists of one unit.
4. The solid oxide fuel cell flow channel structure according to claim 1, characterized in that, The structural dimensions of each branch unit are consistent, and the structural dimensions of each bus unit are consistent.
5. The solid oxide fuel cell flow channel structure according to claim 1, characterized in that, The structural parameters of the flow channel structure are obtained using a multi-objective optimization method based on a surrogate model. The optimization objectives of the multi-objective optimization are to optimize the net power density and oxygen uniformity of the solid oxide fuel cell. The surrogate model is obtained by training an artificial neural network and is used to establish the mapping relationship between the structural parameters and the objective function of the multi-objective optimization.
6. A solid oxide fuel cell, comprising an anode flow channel and a cathode flow channel, characterized in that, The anode flow channel and / or cathode flow channel adopt the solid oxide fuel cell flow channel structure as described in any one of claims 1-5.
7. A multi-objective optimization design method for the flow channel structure of a solid oxide fuel cell as described in any one of claims 1-5, characterized in that, include: Based on the optimization objective, an objective function is established, and optimization variables and their corresponding value ranges are determined. The optimization objective is to optimize the net power density and oxygen uniformity of the solid oxide fuel cell, and the optimization variables are the structural parameters of the flow channel structure. The Latin hypercube sampling method was used to select design points, and the objective function values of each design point were calculated by CFD numerical simulation to form a database. The surrogate model is obtained by training an artificial neural network through the database. The input of the neural network is the optimization variable, and the output is the objective function, namely, the net power density of the battery and the oxygen uniformity. Based on the aforementioned proxy model, a multi-objective optimization algorithm is used to obtain the structural parameters corresponding to the optimal battery net power density and oxygen uniformity.
8. The multi-objective optimization design method for the flow channel structure of a solid oxide fuel cell according to claim 7, characterized in that, Based on the aforementioned surrogate model, a multi-objective genetic algorithm is used to obtain the structural parameters corresponding to the optimal battery net power density and oxygen uniformity, including: Initialize the population, the number of individuals in the population corresponds to the number of groups of the optimization variables, and each group of optimization variables is randomly selected within the range of the values stated; The surrogate model is used to calculate the objective function corresponding to each group of optimization variables, and the objective function is used to evaluate the fitness of individuals within the population. The offspring population is generated through selection, crossover, and mutation; The parent and offspring populations are combined, and the next generation population is formed by non-dominated sorting and crowding calculation. Repeat the above steps until the stopping criterion is met. At the end of the iteration, multiple Pareto nondominated solutions are obtained. The optimal structural parameters are obtained by applying the order preference similarity technique of the ideal solution.
9. The multi-objective optimization design method for the flow channel structure of a solid oxide fuel cell according to claim 7, characterized in that, The optimization variables include: the length of the inlet section of the diversion unit, the length of the converging section of the diversion unit, the height of the outlet of the converging section of the diversion unit, the width of the outlet of the converging section of the diversion unit, and the length of the converging unit.
10. The multi-objective optimization design method for the flow channel structure of a solid oxide fuel cell according to claim 9, characterized in that, The total length of the bus unit and its adjacent branch unit is a set value.