Method for optimizing a fluid flow in a fuel cell by adapting at least one geometric parameter of fluid-guiding elements

Abstract

The invention relates to a method for a computer-implemented optimization of a fluid flow through an inlet section (110), a reaction section (120), and an outlet section (130) of an element of a fuel cell system by adapting at least one geometric parameter of fluid-guiding elements, wherein the following steps are carried out: - carrying out a flow simulation on a simulation model (SM) of the fuel cell (100), - generating a flow map (SK) of the fuel cell (100) on the basis of the results of the flow simulation, - automatically identifying at least one sub-flow (TS) of the fluid flow in the flow map (SK), said sub-flow correlating to at least one fluid-guiding element (140) of the fuel cell (100), on the basis of the results of the flow simulation, - detecting a selection of at least one sub-flow (TS) to be optimized and of the at least one correlating fluid-guiding element (140), - detecting a change in at least one geometric parameter (GP) of the at least one selected fluid-guiding element (140), - repeating the steps of carrying out the flow simulation, generating the flow map (SK), and identifying the at least one sub-flow (TS) as an optimization loop, - generating a comparison result (VE) of the repeated flow simulation with the previous flow simulation, and - outputting the at least one geometric parameter (GP) as an optimized geometric parameter (GP) on the basis of the comparison result (VE) after all of the flow simulations have been completed.

Description

[0001] Optimization method for optimizing a fluid flow in a fuel cell by adapting at least one geometry parameter of fluid guide elements

[0002] The present invention relates to an optimization method for optimizing a fluid flow in a fuel cell system by adapting at least one geometric parameter of fluid guide elements, a computer program product for carrying out such an optimization method, an optimization device for carrying out such an optimization method, and a fuel cell with an optimized fluid guide.

[0003] It is known that the fluid flow in fuel cells can have a decisive influence on the operation and, in particular, the efficiency of the fuel cell. In an individual fuel cell, the fluid is fed via an inlet section to a reaction section where the fuel cell reaction takes place. The reaction products are then discharged from the fuel cell via an outlet section. In each fuel cell, this occurs primarily on two different sides of a fuel cell membrane, which thus separates the anode and cathode sides of the fuel cell.

[0004] A disadvantage of the known solutions is that the design of the fluid flow guidance is largely based on manual or iterative changes based on experience and fluid dynamic principles. Thus, the design is based on fluid dynamic principles with regard to the division, distribution, and guidance of individual partial flows of the fluid flows, particularly in the inlet section and / or the outlet section. Furthermore, it is known to provide guide ribs and / or guide pins in the inlet section and / or the outlet section as fluid guidance elements, which ensure the desired, as even as possible, distribution of the fluid flow to the reaction section. At the same time, however, it must be taken into account that a preferably predetermined pressure loss between the inlet and outlet sections can be maintained as precisely as possible.In the known solutions, this is done by applying empirical values ​​and fluid dynamic principles, so that although it is possible to improve the design of the fluid flow by specifying the geometric parameters of the fluid guide elements accordingly, automatic or semi-automatic optimization is not possible.

[0005] The object of the present invention is to at least partially remedy the disadvantages described above. In particular, the object of the present invention is to improve the fluid flow through a fuel cell in a cost-effective and simple manner by optimizing the geometric parameters of the fluid guide elements.

[0006] The above object is achieved by an optimization method having the features of claim 1, a computer program product having the features of claim 18, an optimization device having the features of claim 16 and a fuel cell having the features of claim 17. Further features and details of the invention emerge from the subclaims, the description and the drawings. Features and details described in connection with the optimization method according to the invention naturally also apply in connection with the computer program product according to the invention, the optimization device according to the invention and a fuel cell according to the invention and vice versa, so that with regard to the disclosure of the individual aspects of the invention, reference is and can always be made reciprocal.

[0007] An optimization method according to the invention serves to optimize a fluid flow through an inlet section, a reaction section, and an outlet section of an element of a fuel cell system by adjusting at least one geometric parameter of fluid guide elements. Such an optimization method is characterized by the following steps, which are continuously repeated with each iteration step, particularly during the simulation run:

[0008] - Carrying out a flow simulation, in particular at least one flow step, on a simulation model of the fuel cell,

[0009] Creating a flow map of the fuel cell based on the results of the flow simulation, in particular of the at least one flow step, - Automated identification of at least one partial flow of the fluid flow, in particular of a target area (for example a reaction area), in the flow map correlating with at least one fluid guide element of the fuel cell based on the results of the flow simulation, in particular of the at least one flow step,

[0010] - detecting a selection, in particular the most relevant, of at least one partial flow to be optimized, in particular a target area, of the at least one correlating guide element,

[0011] - detecting, in particular defining, a change in at least one geometric parameter of the at least one correlating fluid guide element,

[0012] - Repeating the steps of carrying out the flow simulation, in particular the at least one flow step, the creation of the flow map and the identification of the at least one partial flow,

[0013] - generating a comparison result of the repeated flow simulation, in particular of the at least one flow step, with the previous flow simulation, in particular of the at least one flow step,

[0014] - Output of at least one geometry parameter after completion of all flow simulations based on the comparison result as an optimized geometry parameter.

[0015] In particular, this simulation is carried out separately for each medium of the fuel cell (air, fuel, coolant).

[0016] Depending on the manufacturing process of the bipolar plates, all flow spaces can be individually designed and optimized, especially if the two sides of a bipolar plate can be designed differently (e.g., carbon plates). With metal-embossed plates, for example, significantly thinner wall thicknesses can be achieved, but the control elements are the same on both sides (i.e., for two different media!). Since the media supply channels are necessarily different, the control elements must take media flows influenced by one plate (ideally all three) into account during optimization, which makes a manually iterative process all the more complicated.

[0017] An optimization method according to the invention is based on the application of flow simulation. For this purpose, a digital simulation model is created for a defined, predefined geometry of a fuel cell, and in particular for the fluid flow elements present in the inlet and / or outlet sections. This model represents the actual media sections for the flow simulation. This model is, in particular, a 3D model that can be used as the basis for computer-based simulation software. Flow simulation can also be referred to as CFD (Computational Fluid Dynamics) simulation and can be implemented using a wide variety of software solutions. This step of flow simulation can also be understood as the numerical and physical convergence of the initial state.

[0018] It should also be noted that, in an optimization method according to the invention, the step of performing a flow simulation can, in particular, also be implemented as a single run of such a simulation. In such a case, this step could also be referred to as performing a flow step of an iterative flow simulation.

[0019] As a result of the flow simulation, the optimization process in the next step obtains a flow map (division of the entire flow area into flow zones) of the fuel cell, which in particular determines the individual flow lines, preferably with a width extension in the form of flow bands of the fluid flow, and / or the pressure conditions along the distribution across the fuel cell. One possibility is to simulate an additional transport equation, which, similar to coloring a segment in the supply area, can color-code the entire subsequently affected flow path from this segment, thereby also taking mixing and dilution effects into account. The flow map thus represents the simulation result, which subsequently forms the basis for the optimization steps for optimizing the geometry of the fluid guide elements.

[0020] Based on the created flow map, it is now possible according to the invention to automatically identify at least one partial flow with respect to a software product used, such that it correlates with at least one fluid guide element of the fuel cell. In other words, it becomes possible to recognize streamlines or flow paths of the fluid flow through the fuel cell as partial flows, so that a direct correlation of the flow influence of the individual fluid guide elements on the respective partial flow can be identified. In other words, the relationship is established between fluid guide elements, on the one hand, and the partial flow directly influenced by these fluid guide elements, on the other. This identification is carried out on the basis of the automatic optimization method and, in particular, is free from manual intervention.The results of the flow simulation in the form of the flow map serve to perform this identification step, which is based in particular on the geometric correlation, the geometric parameters of the fluid flow elements in the initial state of the simulation model, and the extent of the individual partial flow lines and / or partial flow paths. Individual fluid flow elements can have significantly different distances from the respective partial flow.

[0021] As soon as the correlation between the individual partial flows and the individual fluid guide elements has been identified, in particular automatically, a selection of at least one partial flow to be optimized is recorded. This recording can be based on manual input by a user, but also, as explained later, automated or semi-automated. In other words, the focus for the subsequent optimization loop is directed at at least one partial flow. The complex optimization problem of all sections and the entire flow can thus be reduced to the at least one partial flow. When the optimization process is run on an executing computer, a selection of at least one partial flow to be optimized is recorded and, based on the preceding automatic identification, the assignment to the respective correlating fluid guide element is determined.Preferably, consideration is given here to the target area most affected by a change in a guide element, which is located downstream and can also be further influenced by neighboring and subsequent guide elements and thus by mixing effects. This assignment can change accordingly with each iteration during the simulation. Such a change occurs particularly in the initial phase of such an optimization process. Compared to a purely automatic optimization, this recording step can focus the optimization on at least one partial flow of increased interest, thus reducing the computational effort until the optimization goal is achieved in the form of the optimized geometric parameters. This allows at least one change in a geometric parameter to be recorded for this selected correlating fluid guide element in the next step.The geometry parameters can also be changed by manual input from a user or by automated or semi-automated specification as part of an automatic optimization process. This clearly shows another advantage of the focused optimization of the present invention. This allows the fluid guide element that is relevant to the partial flow to be optimized because it has an influence to be directly selected and its geometry parameters modified. This further reduces the number of necessary optimization iterations and the associated computing time. As soon as a capture, whether manually or automatically specified, has been performed, the selected fluid guide element is present in a first modified state in the three-dimensional simulation model and thus serves as the starting point for a further simulation as an optimization loop.

[0022] In order to examine precisely how this change in this geometric parameter of this at least one fluid guide element has a positive or negative influence on the fluid flow through the fuel cell, the steps of performing the flow simulation, creating the flow map, and identifying the at least one partial flow are repeated. However, the focusing already explained already ensures that the change implemented is also relevant for the focused partial flow. In other words, an optimization loop can be run here, which allows the effect of this change in the at least one changed geometric parameter on the fluid flows to be identified based on a changed geometric parameter.To evaluate this impact, a comparison result of the repeated flow simulation is generated with the results of the previous flow simulation. Comparison with further, even more recent flow simulations is also possible here, as explained later with regard to gradient formation.

[0023] It is clearly evident here that the optimization problem can now be reduced to changing the geometry parameters. Optimization with regard to the correlation between the fluid guidance elements and their influence is not necessary, since the inventive focus on at least one partial flow already specifies this correlation in a defined manner. Compared to a general optimization method, the time and computational effort required to achieve the optimized geometry parameters are thus significantly reduced. The comparison result now allows a statement to be made, particularly qualitatively and / or quantitatively, as to whether the implemented and recorded change to the geometry parameter has led to an improvement and / or a deterioration of this partial flow.For example, the comparison result can be automatically compared with specified values ​​that deal in particular with flow parameters such as flow velocity, flow direction, flow pressure or similar.

[0024] At the end of an optimization method according to the invention, at least one geometry parameter is output based on the comparison result. Of course, multiple geometry parameters can also be output, preferably together with the respective comparison result, so that a selection of changes with the corresponding effect becomes visible to the user of such an optimization method. Preferably, in particular, the geometry parameter that led to the best comparison result from the perspective of the specified values ​​explained later is output as the optimized geometry parameter. Such a geometry parameter is referred to as an optimized geometry parameter, so that an optimization method according to the invention preferably outputs at least one optimized geometry parameter as a result.

[0025] A method according to the invention thus uses flow simulation to optimize the design of media guides in a fuel cell system in a focused manner by geometrically modifying fluid guide elements. The starting point and thus input to the flow simulation is the digital representation of a real media guide design, for example, a bipolar plate of a fuel cell system. The output of the focused optimization is an optimized, modified design of the media guide with regard to new geometries of the fluid guide elements.These optimized design parameters are now achieved through focused optimization with less optimization time, more precisely, and with less computational effort. It can be advantageous if, in an optimization method according to the invention, when generating the comparison result, at least one default value, for example a mass flow, a temperature, a media concentration, or the like, is taken into account for an evaluation of the change in at least one geometry parameter, wherein this evaluation is taken into account for the subsequent detection of a change in the geometry parameter. Such a default value can, for example, relate to flow parameters of the fluid flow and / or the partial flow. For example, the flow velocity, the flow direction, the flow cross sections of the respective partial flow, the flow pressure, or the like can be used as default values.In particular, when the flow map relates to a resulting pressure loss across the surface of the fuel cell, the pressure of the respective partial flow section can represent a preset value within the meaning of the present invention. In other words, it becomes possible to compare, with respect to a preset value of a desired or to-be-achieved pressure loss, whether the implemented change to the geometry parameter has reached the desired preset value or has at least moved toward this desired preset value. This makes it possible, based on this comparison result, to propose a further change to the geometry parameter or even to implement it automatically, particularly in an automated or semi-automated manner.If the comparison result deteriorates compared to the specified value, the implemented change had a negative and therefore undesirable effect, so the next change reverses the first change and / or a change is made in a different or even opposite direction. If the comparison result improves with respect to the specified value, the implemented and recorded change to the geometry parameter has been initially successful, so the next change to the geometry parameter can be made in the same way and / or in the same direction. This, too, is preferably carried out semi-automatically or automatically for the change and thus for the subsequent recording of the change.

[0026] It is advantageous if one or more passive scalar equations are used to identify a correlation between the partial flow of the fluid flow and the at least one fluid guide element. This allows the fluid flow to be optimized even better and more efficiently because the correlation can be reliably detected. It is also advantageous if, in an optimization method according to the invention, the steps of detecting a change in a geometric parameter, repeating the simulation, creating it, and identifying it form an iteration of an optimization loop, wherein a plurality of iterations are carried out. The iteration loop, which can also be referred to as an optimization loop, thus serves to detect a plurality of, in particular, small changes in at least one geometric parameter and to generate a qualitative and / or quantitative comparison result with each run through the optimization loop.In other words, after each iteration, a comparison result is available that contains information about whether the change in the geometry parameter applied and underlying this iteration had a positive or negative effect, focusing on the partial flow. This allows for multiple iterations to evaluate a multitude of different changes to the geometry parameter and, based on the best comparison result, to identify and ultimately output the corresponding best and thus optimized geometry parameter as the optimization result. However, all of these iterations already focus on at least one partial flow and the associated fluid flow elements.

[0027] In an optimization method according to the preceding paragraph, it may be advantageous if at least one termination criterion for terminating the iteration is specified, in particular one of the following:

[0028] - minimal change in the comparison result,

[0029] - reaching a target value,

[0030] - Reaching a maximum number of iterations.

[0031] The above list is not exhaustive. A minimal change in the comparison result means that the change in the geometry parameter performed, which is the basis for the comparison result, has only a minimal effect on the flow simulation being performed. In other words, a further change to this direction or fluid guide element may no longer be successful and thus have no significant impact on the fluid flow. The iteration can be aborted at this point. Even if a specified value, for example, a desired pressure loss in a specific area of ​​the fuel cell, is reached, this can serve as a termination criterion.To prevent an optimization procedure from exceeding a predefined time limit, the number of iterations or the duration of the total computing time can also be limited and represent a corresponding technical termination criterion for the optimization procedure.

[0032] It is further advantageous if, in an optimization method according to the invention, the comparison results and the recorded changes in the geometry parameters are stored for at least two iterations, with a gradient for the iteration being determined on the basis of these changes. In this way, it becomes possible to identify a trend across multiple changes and multiple iterations, indicating which direction of change and / or which quantity of change can be expected to have which effect in the flow simulation on the respective partial flow. This allows at least a partial prediction to be made, so that the change in the geometry parameter can be carried out in a targeted, automatic and / or semi-automatic manner based on this gradient information, and thus the optimization can proceed more specifically and therefore more quickly.The reduction in optimization effort achieved through focusing until the optimization goal is achieved is further improved in this way.

[0033] It is also advantageous if, in an optimization method according to the invention, geometric limits for the respective fluid guide element are taken into account when detecting a change in the geometric parameters. In the actual arrangement of fluid guide elements, in particular in the inlet section and / or the outlet section, these cannot be arbitrarily enlarged, reduced and / or shifted. For example, geometric limits exist in such a way that the position and extent of adjacent fluid guide elements must be taken into account, since otherwise the partial flow would be blocked or components would overlap. It may also be the case that a geometric limit exists in such a way that the minimum free flow cross-section in the region of the fluid guide element must not be undercut in order to prevent the total volume flow supplied to the reaction section of the fuel cell from being undesirably undercut.Furthermore, there are also limitations regarding the manufacturability of the small size of the finally determined fluid guide element and also of the intermediate regions corresponding to the shaping tools. Furthermore, it is also advantageous if, in an optimization method according to the invention, the change to at least one geometric parameter takes place automatically, in particular based on the generated comparison result. As already explained, various information from the iteration can be used to evaluate the success or failure of the implemented change. Based on this evaluation of the comparison result, the optimization method itself, when executed on a computer, can now carry out an automated change preferably in a favored direction, i.e. in the direction of an expected successful influencing of the partial flow.This can be done fully automatically, but also semi-automatically as a suggestion, which must be manually approved and accepted by the software operator. The amount of change to the geometry parameters is preferably determined individually for each guide element, essentially based on the deviation of the actual variable from the target variable (mass flow, temperature, media concentration, etc.). Additional criteria include the values ​​of a passive scalar, the distance of the respective guide element from the target / reaction area, and / or the force / moment of the flow acting on the respective guide element.

[0034] A further advantage can also be achieved if, in an optimization method according to the invention, a selection of at least two partial flows is recorded for at least two different correlating fluid guide elements, in particular for all fluid guide elements of the inlet section, the reaction section, and / or the outlet section of the fuel cell. While an optimization method according to the invention already provides the described advantages for the optimization of a single fluid guide element, a large number of fluid guide elements are typically provided in real fuel cells.If the optimization method according to the invention is now applied to additional, in particular all, fluid guide elements, the required optimization time in the form of computing time increases significantly in the embodiment on a computer. However, an overall optimization of all partial flows and thus of the entire fluid flow through the fuel cell can be achieved in this way. In particular, cross-influences between the individual fluid guide elements on one another are also taken into account in this way, since the existing correlation between partial flows and fluid guide elements allows the fluid guide elements that have the maximum influence on the corresponding partial flow to be identified. This identification, in particular, represents a decisive advantage of the present optimization method when applied to multiple fluid guide elements.These complex relationships and the combined influence of individual fluid guidance elements can only be identified by automation in the optimization method according to the invention and, due to their complexity, are not accessible to manual evaluation based on fluid dynamic principles or are only accessible to a very limited extent.

[0035] Further advantages are also achieved if, in an optimization method according to the invention, the partial flows for different fluid guide elements can overlap at least in sections. This means that the partial flows receive their influencing factors from different fluid guide elements. It is therefore possible for different fluid guide elements to exert a joint influence on a section of a partial flow that overlaps with other partial flows. Here, too, it is clearly evident how the complexity increases further due to the overlap of the partial flows and the corresponding overlapping influences of the fluid guide elements, whereby the corresponding high-quality optimization and mutual dependence can only be achieved by an optimization method according to the invention and only to a limited extent by manual fluid optimization.

[0036] It is also advantageous if, in an optimization method according to the invention, only at least one geometry parameter of at least one fluid guide element is changed in an iteration consisting of the detection of a change in a geometry parameter and the repetition of the simulation, creation, and identification for each iteration. Of course, when applied to multiple fluid guide elements, each individual fluid guide element can also be used and optimized multiple times, so that an overall optimization of all fluid guide elements of the inlet section, the reaction section, and / or the outlet section can take place. Of course, however, in principle two or more geometry parameters of exactly one fluid guide element can also be adjusted per iteration. In particular, all guide elements should be checked to determine whether a change is necessary.Only then can holistic optimization be achieved. However, the change is implemented in very small dimensions and in detailed steps to ensure that the effects of the many individual control elements are not overly pronounced, which is why distance- and concentration-dependent damping or reduction factors are applied.

[0037] It is also advantageous if, in an optimization method according to the invention, the fluid guide element has at least one of the following configurations:

[0038] - guide channel,

[0039] - guide rib,

[0040] - guide surface,

[0041] - Guide pin.

[0042] The above list is not exhaustive. In particular, different designs of fluid guide elements can be combined with one another in a fuel cell. Accordingly, different geometric parameters, such as extensions, thicknesses, positioning, diameters, shapes, curvatures, or the like, can be provided for each type of fluid guide element, whereby the respective geometric parameters to be changed can be designed specifically for the respective fluid guide element.

[0043] Furthermore, it can be advantageous if, in an optimization method according to the invention, the identified at least one partial flow is smaller than the total flow of the fluid flow in the flow map. This means that the partial flow always represents a reduction in the volume on which the subsequent optimization for this partial flow is based. This step further clarifies the previously discussed focusing, since further optimization only needs to be performed for the reduced volume. In particular, an identification and / or selection of exactly one partial flow for the optimization method takes place.

[0044] It is also advantageous if, in an optimization method according to the invention, when the flow simulation is repeated, it is only carried out for the section of the fuel cell simulation model that has the identified at least one partial flow. Here, the focusing step is further strengthened, since the computational effort for the optimization iteration can be significantly reduced. Due to the usual design of flow simulations, the computational effort for the implementation increases with the size of the volume under consideration and with the accuracy of the simulation. By reducing the volume under consideration by means of the reduction to the selected partial flow, either the computational effort can be reduced while maintaining the same simulation accuracy and / or the simulation accuracy can even be increased.

[0045] Furthermore, it can be advantageous if, in an optimization method according to the invention, the detection of the partial flow to be optimized comprises an automatic selection, wherein at least one partial flow is automatically selected for detection which differs from neighboring partial flows and / or from an average value for the at least one flow parameter for the overall flow of the flow map with regard to at least one flow parameter. As already explained, a manual selection can be recorded via the detection interface. In the present embodiment, however, this also occurs automatically, thereby further increasing independence from the user's experience. In fuel cells, flow fronts in the fuel cells that are as uniform as possible are particularly advantageous.Therefore, a partial flow that deviates from such uniformity, for example, by a predefined value, can represent a reasonable selection for optimization. In other words, this automatic selection can now specifically promote flow uniformity, so that in physical reality, the fuel cell develops a more uniform fluid flow with the optimized geometric parameters of the fluid guide elements correlated with this partial flow.

[0046] A further subject matter of the present invention is a computer program product comprising instructions which, when executed by a computer, cause the computer to carry out the steps of an optimization method according to the present invention. Thus, a computer program product according to the invention provides the same advantages as have been explained in detail with reference to an optimization method according to the invention. The present invention also relates to an optimization device for carrying out an optimization method according to the invention. Such an optimization device is characterized by a simulation module for performing a flow simulation on a simulation model of the fuel cell and for creating a flow map of the fuel cell based on the results of the flow simulation.Furthermore, the optimization device is equipped with an identification module for automated identification of at least one partial flow of the fluid flow in the flow map correlating with at least one fluid guide element of the fuel cell based on the results of the flow simulation.

[0047] With the aid of a detection module, a selection of at least one partial flow to be optimized and the at least one correlating guide element are detected, as well as a change in at least one geometric parameter of the at least one selected fluid guide element is detected. A comparison module serves to generate a comparison result of the repeated flow simulation with the previous one, in particular in the form of an individual flow step, wherein an output module subsequently outputs the at least one geometric parameter after completion of all flow simulations, in particular all flow steps, based on the comparison module as an optimized geometric parameter. The simulation module, the identification module, the detection module, the comparison module, and / or the output module are designed in particular for carrying out a method according to the invention.Thus, an optimization device according to the invention provides the same advantages as those explained exclusively with reference to an optimization method according to the invention. For this purpose, the optimization device comprises, in particular, a computing unit, for example in the form of a computer, to execute the individual method steps and to form the described modules.

[0048] A further subject of the present invention is a fuel cell for generating electrical power and / or conducting electrolysis. Such a fuel cell has an inlet section, a reaction section, and an outlet section for conducting a fluid. The fuel cell is characterized in that at least one fluid guide element is arranged in the inlet section, the reaction section, and / or the outlet section, which element has at least one geometric parameter optimized using an optimization method according to the invention.

[0049] Further advantages, features, and details of the invention will become apparent from the following description, in which exemplary embodiments of the invention are described in detail with reference to the drawings. They show schematically:

[0050] Fig. 1 shows an embodiment of an optimization device according to the invention,

[0051] Fig. 2 shows an embodiment of a fuel cell according to the invention,

[0052] Fig. 3 shows a flow map,

[0053] Fig. 4 a possible recorded selection of a partial flow,

[0054] Fig. 5 an alternative possible selection of a detected partial flow,

[0055] Fig. 6 a possible detected change of a geometry parameter,

[0056] Fig. 7 shows another possible detected change of a geometry parameter and

[0057] Fig. 8 shows an alternative embodiment of an optimization device according to the invention.

[0058] Figure 1 schematically shows an optimization device 10 capable of implementing an optimization method according to the invention. A simulation model SM is provided as the basis for implementing the optimization method, which provides, for example, the geometric extensions and guidance conditions for fluid flows in three-dimensional form, particularly in the form of CAD data.

[0059] Figure 2 shows a possible geometric design of such a fuel cell. This fuel cell 100, shown here from one side, for example the anode side, is provided with an inlet section 110, a reaction section 120, and an outlet section 130. A plurality of guide pins are provided in the inlet section 110 and the outlet section 130 as fluid guide elements 140, which enable the inlet fluid flow to be divided and the outlet fluid flow to be collected. Furthermore, the connection to the inlet section 110 is divided into a plurality of inlet subsections ET, which can later be assigned to individual subflows TS in the form of passive scalars. Similarly, an outlet subsection AT is also provided at the outlet section 130.

[0060] In Figure 1, a flow simulation is now carried out in the simulation module 20 based on the CAD data of the fuel cell 100. The result of this flow simulation is, in particular, a flow map SK, as also shown in more detail in Figure 3. In this embodiment, the flow map SK is shown in the form of different pressure levels, designated here by the Roman numerals I to VII. In other words, the flow gradually loses pressure from the inlet sections ET to the outlet sections AT, so that a pressure drop extends over the entire surface of the fuel cell 100 in the manner shown. The formation of the individual pressure fronts, as can be seen in the flow map SK in Figure 3, depends on the actual fluid flow, in particular the individual partial flows TS, which will be explained later.

[0061] As Figure 1 shows, this flow map SK is now fed to the identification module 30 as a starting point. The identification module 30 is now able to identify at least one partial flow TS. For example, Figure 4 shows an identified partial flow TS, which here correlates with an associated inlet section ET. This partial flow TS in the identification according to the embodiment of Figure 4 now correlates with a small selection of fluid guide elements 140, here six individual guide pins of the guide elements 140. This correlation can be further specified for each guide pin of the guide elements 140 that correlate with this identified partial flow TS. In other words, it is now possible in the detection module 40 to detect the identified partial flow TS and the correspondingly correlated selected fluid guide elements 140.

[0062] Secondly, the detection module 40 serves to detect a change in a geometry parameter GP. Figure 6 shows an example of this, so that, for example, by moving the guide pin as the fluid guide element 140 while maintaining its diameter, this displacement can be understood as a change in a geometry parameter GP. This change in the geometry parameter GP can be performed purely manually by the user, but is particularly automated or semi-automated, as has already been explained several times. Shape-changing measures or elements with multiple shape parameters can also be created or modified.

[0063] Figure 1 now shows how, after the acquisition of the changed geometry parameter GP and the selected fluid guide element 140, a repetition of the flow simulation and the identification of the partial flows TS is performed. This should be represented as an internal repetition loop, as the same components of a computer device as those used at the start of the optimization method according to the invention.

[0064] After the run, a termination criterion is also queried in a comparison module 50, so that finally in the output module 60 at least one individual geometry parameter GP, in particular an optimized geometry parameter GP, can be output.

[0065] Also in Figure 1, as well as Figure 8, a particularly simple variant is shown with a dash-dot line, in which a direct feedback is provided into the simulation model SM.

[0066] Figure 5 shows an alternative embodiment of a selection of a partial flow TS. This is based on the same flow map SK as the selection in Figure 4, namely the flow map SK according to Figure 3. Here, however, the partial flow TS also relates to a different partial inlet section ET in the identified manner, so that a smaller and different selection of fluid guide elements 140 is also correlated with this partial flow TS and has thus been identified as influencing fluid guide elements 140. Here, a single fluid guide element 140 can again be selected and recorded with regard to a change in a geometry parameter GP, so that the optimization can also be carried out for this partial flow TS in the manner described.

[0067] Figure 7 shows an alternative embodiment of a change in a geometry parameter GP. Here, the positioning of the fluid guide element 140 remains the same, but its extension changes—here, the diameter of the guide pin as the geometry parameter GP. Of course, different geometry parameters GP and different types of changes to the geometry parameters GP can also be used in an optimization method according to the invention.

[0068] Figure 8 shows another further development of the optimization device 10 of Figure 1. Thus, an iteration loop IT is shown here in an iterative manner, which shows that over a plurality of iterations, one or more fluid guide elements 140 for one or more partial flows TS can run through the optimization process, so that finally, one or more optimized geometry parameters GP can be output from the output module 60 for one or more fluid guide elements 140. Preferably, for one or more geometry parameters GP, reference is also made here to one or more default values ​​VW, which make the comparison result VE quantitatively and / or qualitatively assessable. Preferably, however, all partial flows TS are recorded in parallel, and the geometry parameters GP are also determined.The changes are imposed in a significantly damped manner in order to achieve a step-by-step convergence of the entire flow space and to collectively take into account the respective interaction of all influences.

[0069] The above explanation of the embodiments describes the present invention exclusively by way of examples.

[0070] List of reference symbols

[0071] 10 Optimization device

[0072] 20 Simulation module

[0073] 30 Identification module

[0074] 40 Comparison module

[0075] 50 recording module

[0076] 60 Output module

[0077] 100 fuel cells

[0078] 110 Inlet section

[0079] 120 Reaction section

[0080] 130 outlet section

[0081] 140 Fluid guide element

[0082] SM simulation model

[0083] SK flow map

[0084] TS partial flow

[0085] GP geometry parameters

[0086] VE comparison result

[0087] VW default value

[0088] IT Iteration

[0089] ET inlet section

[0090] AT outlet section

[0091] I - VII pressure levels

Claims

Patent claims 1. for a computer-implemented optimization of a fluid flow through an inlet section (110), a reaction section (120) and an outlet section (130) of an element of a fuel cell system by adjusting at least one geometric parameter (GP) of fluid guide elements (140), characterized by the following steps: - carrying out a flow simulation on a simulation model (SM) of the fuel cell (100), - Creating a flow map (SK) of the fuel cell (100) based on the results of the flow simulation, - Automated identification of at least one partial flow (TS) of the fluid flow in the flow map (SK) correlating with at least one fluid guide element (140) of the fuel cell (100) on the basis of the results of the flow simulation, - detecting a selection of at least one partial flow (TS) to be optimized and the at least one correlating fluid guide element (140), - detecting a change in at least one geometry parameter (GP) of the at least one selected fluid guide element (140), - Repeating the steps of performing the flow simulation, creating the flow map (SK) and identifying at least one partial flow (TS) as an optimization loop, - Generating a comparison result (VE) of the repeated flow simulation with the previous flow simulation, - Output of at least one geometry parameter (GP) after completion of all flow simulations based on the comparison result (VE) as an optimized geometry parameter (GP).

2. Optimization method according to claim 1, characterized in that when generating the comparison result (VE) at least one default value (VW) is taken into account for an evaluation of the change in the at least one geometry parameter (GP), wherein this evaluation is taken into account for the subsequent detection of a change in the geometry parameter (GP).

3. Optimization method according to claim 1 or 2, characterized in that one or more passive scalar equations are used to identify a correlation of the partial flow (TS) of the fluid flow and the at least one fluid guide element (140).

4. Optimization method according to one of the preceding claims, characterized in that the steps of detecting a change in a geometry parameter (GP), repeating the simulation, creating and identifying form an iteration (IT) of an optimization loop, wherein a plurality of iterations (IT) are carried out.

5. Optimization method according to claim 4, characterized in that at least one termination criterion for terminating the iteration (IT) is specified, in particular one of the following: - Minimal change in the comparison result (VE) - Reaching a target value (VW) - Reaching a maximum number of iterations (IT) 6. Optimization method according to one of claims 4 or 5, characterized in that the comparison results (VE) and the detected changes in the geometry parameters (GP) are stored for at least two iterations (IT), on the basis of which a gradient for the iteration (IT) is determined.

7. Optimization method according to one of the preceding claims, characterized in that when detecting a change in the geometry parameters (GP), geometry limits for the respective fluid guide element (140) are taken into account.

8. Optimization method according to one of the preceding claims, characterized in that the change of the at least one geometry parameter (GP) takes place automatically, in particular on the basis of the generated comparison result (VE).

9. Optimization method according to one of the preceding claims, characterized in that a selection of at least two partial flows (TS) is recorded for at least two different correlating fluid guide elements (140), in particular for all fluid guide elements (140) of the inlet section (110), the reaction section (120) and / or the outlet section (130) of the fuel cell (100).

10. Optimization method according to claim 9, characterized in that the partial flows (TS) for different fluid guide elements (140) can overlap at least in sections.

11. Optimization method according to one of claims 9 or 10, characterized in that in an iteration (IT) from the detection of a change in a geometry parameter (GP) and the repetition of the simulation, the creation and the identification for each iteration (IT) only at least one geometry parameter (GP) of exactly one fluid guide element (140) is changed.

12. Optimization method according to one of the preceding claims, characterized in that the fluid guide element (140) has at least one of the following configurations: - Guide channel - Guide rib - Guide surface - Guide pin 13. Optimization method according to one of the preceding claims, characterized in that the identified at least one partial flow (TS) is smaller than the total flow of the fluid flow in the flow map (SK).

14. Optimization method according to one of the preceding claims, characterized in that when repeating the flow simulation, it is only carried out for the section of the simulation model (SM) of the fuel cell (100) which has the identified at least one partial flow (TS).

15. Optimization method according to one of the preceding claims, characterized in that the detection of the partial flow (TS) to be optimized comprises an automatic selection, wherein at least one partial flow (TS) is automatically selected for detection which differs with respect to at least one flow parameter from adjacent partial flows (TS) and / or from an average value for the at least one flow parameter for the overall flow of the flow map (SK).

16. Optimization device (10) for carrying out an optimization method, characterized by a simulation module (20) for carrying out a flow simulation on a simulation model (SM) of the fuel cell (100) and creating a flow map (SK) of the fuel cell (100) based on the results of the flow simulation, an identification module (30) for automatically identifying at least one partial flow (TS) of the fluid flow in the flow map (SK) correlating with at least one fluid guide element (140) of the fuel cell (100) based on the results of the flow simulation, a detection module (40) for detecting a selection of at least one partial flow (TS) to be optimized and the at least one correlating fluid guide element (140) and detecting a change in at least one geometric parameter (GP) of the at least one selected fluid guide element (140),a comparison module (50) for generating a comparison result (VE) of the repeated flow simulation with the previous flow simulation, and an output module (60) for outputting the at least one geometry parameter (GP) after completion of all flow simulations on the basis of the comparison result (VE) as an optimized geometry parameter (GP), wherein the simulation module (20), the identification module (30), the detection module (40), the comparison module (50) and / or the output module (60) are designed for executing a method having the features of one of claims 1 to 15.

17. Fuel cell (100) for generating electrical current and / or carrying out electrolysis, comprising an inlet section (110), a reaction section (120) and an outlet section (130) for guiding a fluid, characterized in that at least one fluid guiding element (140) is arranged in the inlet section (110), in the reaction section (120) and / or in the outlet section (130) with at least one geometry parameter (GP) optimized using an optimization method having the features of one of claims 1 to 15.

18. A computer program product comprising instructions which, when executed by a computer, cause the computer to carry out the steps of an optimization method having the features of one of claims 1 to 15.

Images