DE102009036197A1 - Method for operating fuel cell system in e.g. motor vehicle, involves measuring input electrical voltage applied to fuel cell, and controlling valve mechanism partially depending on measured electrical voltage - Google Patents

Abstract

The method involves supplying hydrogen to an anode region (3) of a fuel cell (2). Water and/or gas are discharged into an area of an anode circuit (8) over a valve mechanism (12). The discharged water and/or gas are brought into a volumetric flow flowing to a cathode region (4) of the fuel cell. An input electrical voltage applied to the fuel cell is temporally measured. The valve mechanism is partially controlled depending on the measured electrical voltage. The valve mechanism is opened in dependence of a computed quantity of water in the anode circuit.

Description

The The invention relates to a method for operating a fuel cell system after further defined in the preamble of claim 1 Art.

Fuel cell systems As is known, at least one fuel cell which usually consists of several individual cells, the are combined into a fuel cell stack or stack. The fuel cell has an anode space and a cathode space and have a membrane disposed therebetween, for example an ion-conducting membrane made of a polymer electrolyte (PEM membrane), on. So-called PEM fuel cells must be fueled be, which has a certain moisture to high efficiency to reach and existing in the fuel cell membranes wet to keep and thereby avoid damage with not sufficiently humidified membranes can occur. Of the Fuel is usually hydrogen gas. That at the For example, product water resulting from fuel cell reaction collected in a water separator and can be used for humidification become. It is also known that in a supply the fuel cell unit with pure hydrogen gas with recirculation the unused hydrogen in the exhaust gas in an anode circuit to the anode inlet, the proportion of nitrogen and water in the anode circuit gradually increases and deteriorates the efficiency the fuel cell unit leads. This is prevented by either a portion of the flowing gases continuously drained or a bleed valve open discontinuously is from time to time with the anode-side flow circuit with to purge fresh hydrogen and increase the efficiency keep this way at a height level by the Hydrogen concentration in the feedback loop set and inert gases from the feedback loop be removed. Through this flushing operation (purges) succeed it to significantly stabilize the performance of the fuel cell unit.

From the DE 699 12 918 T2 a fuel cell system is known in which a return of the hydrogen is provided. From the anode circuit branch off a first and a second purge line, which are provided during the flushing process mainly for the discharge of water from a Wasserabscheidevorrichtung and nitrogen. Conventionally, in known systems, one valve (purge valve and drain valve) and one line, including any necessary filters, are required for flushing with hydrogen (purge) and for removing water (drain), with separated liquid water being removed from the recirculation circuit during water removal. The opening and closing of the purge valve is controlled either via a hydrogen concentration measurement or it is time-controlled opened and closed depending on the load. The purge line ends, depending on the concept, in the cathode output or in the cathode input. The opening and closing of the drain valve is usually via a level sensor, which indicates an upper (opening of the valve) and a lower (closing of the valve) level in the Wasserabscheidevorrichtung. The drain usually terminates at the cathode output of the system.

Next the effort to provide these components in the system cause the components further effort in terms of their control and a possibly necessary for this sensor. Furthermore have appropriate line elements of the valve devices into the respective areas in which the media are being broadcast, to be available. This requires corresponding components and corresponding Space. Even at temperatures below freezing a Safe start and secure functionality of the system to be able to guarantee these Conduit elements also isolated and / or accordingly be formed heated. This also increases the effort in terms of cost, complexity and weight in the above illustrated fuel cell system.

From the international application WO 2008/052578 A1 Also known is a fuel cell system having an anode circuit, referred to herein as a fuel circuit. The special feature of this design is that the functionality of the water separator with a drain valve for discharging the water and the functionality of the blow-off valve for blowing off the nitrogen-containing gas are combined. The structure provides that a water separator is provided with a corresponding valve device. Whenever a correspondingly large amount of water has accumulated, it is drained from the water separator via the valve device. In addition, after the water is drained, gas exits the anode circuit via the valve means of the water separator before it is closed again. The functionality, which was distributed in the above-mentioned writing on two separate components, is thereby integrated into a single component.

This structure already represents a significant improvement over the above-mentioned construction. However, a corresponding sensor for detecting the water level in the separator is also necessary here. Experience in practice has shown that this sensor is extremely easily contaminated with deposits from the fuel cell system, and that this very often leads to a malfunction of the sensor and thus to a frequent or too rare draining the accumulating water leads. This is correspondingly problematic in terms of the reliability of such a system. In addition, the use of such a sensor always also represents a certain expense, since in particular a level sensor in the water separator must be typically designed in the form of two individual sensors, which in turn is associated with the corresponding costs.

From the further general state of the art is also the DE 699 20 279 T2 known. This document deals with the "disposal" of anode exhaust gas in the area of the cathode, without using an anode circuit. In principle, however, it is known from this document that, as stated in Section 16 of the cited document, the metering of hydrogen-containing gas into the region of the cathode causes only a minimal change in the cell voltage and the cell temperature.

task The present invention is now a method of operation to provide a fuel cell system with an anode circuit, which can be constructed easily and efficiently, and in which a safe and reliable operation even without the use complicated sensor technology is made possible.

According to the invention this task by the method with the in the characterizing part solved by claim 1 mentioned features. The dependent ones Claims describe advantageous embodiments and Further developments of the method according to the invention.

Across from the comments in the above to the further prior art mentioned writing has become more unexpected and surprising to the inventors Way shown that the entry of hydrogen in the range of Cathode has a strong influence on the cell voltage. Corresponding studies have shown that the cell voltage even with small amounts of hydrogen in the cathode, in particular, amounts below 10%, with a slump in tension in the fuel cell. This effect of the voltage dip, as soon as hydrogen appears on the cathode side of the fuel cell, was not directly clarified in the investigations. A first guess that this with the reduced air stoichiometry could not be true, as well as experiments with strongly superstoichiometric Conditions have shown the voltage dip. The meanwhile suspected reason lies in a lowering of tension due to a Reduction of the cathode potential by the addition of hydrogen.

In In any case, this voltage drop occurs in all operating situations the fuel cell is always on when hydrogen enters the area the cathode or the cathode compartment passes. Order now at a usual Fuel cell system with anode circuit draining water and control inert gases from the anode cycle, it is possible in the introduction of these two media in the range of Kathodenzuluft about to detect the voltage dip that hydrogen is delivering in the cathode arrives. At this time, therefore, the inert gases and / or the Drained water accordingly and the valve device for draining This media can be closed again. The invention Procedure uses this voltage dip to control the Drains and / or Purges, so on more sensors, such as Water level sensors in the area of a water separator or hydrogen sensors can be dispensed with in the field of cathode supply. The procedure makes it possible, a corresponding fuel cell system much easier and cheaper to build, because now only detected a break in the already measured voltage must be to determine that the drain or the purge accordingly can be stopped. This allows for a very simple and inexpensive construction, which also independent possibly fault-prone sensors works and therefore works safely and reliably.

In a particularly favorable and advantageous embodiment of the method according to the invention, it is provided that after opening the valve device is waited for a maximum predetermined time to a voltage dip, and if this does not occur, the valve device is closed again. If no voltage drop occurs in the area of the fuel cell itself when the valve device is open, then this is a sign that no hydrogen arrives in the region of the anode. In principle, as already explained above, this can mean that the purge, ie the discharge of water and inert gas from the anode region, continues to take place. If a corresponding maximum time is waited for, so that it is no longer to be expected that inert gas and water from the region of the anode will arrive via the valve device in the cathode, then the voltage drop which has still not occurred speaks for an error in the area of the valve device. Such an error may be, for example, a lack of functionality of the valve device. In particular, after the start of a fuel cell system at temperatures below freezing, this may also mean that water was present in the region of the valve device or lines, which is now frozen and impeded the connection to the cathode accordingly. The absence of tension Burglary can also be used according to the inventive idea to check the functionality of the valve devices and the line elements.

In a particularly favorable and advantageous development From this it is therefore envisaged that closing the Valve device without voltage drop generates an error signal. about This error signal can then be detected that there is a problem. This can for example be a retry after expiration of a predetermined time or even a maintenance of the system.

In a particularly favorable development of this is it then provided that the fuel cell system during the presence of an error signal with increased anode pressure is operated. If such an error signal is present, for example due to a frozen pipe, a defect in the valve or Likewise no purge from the anode area occurred. In order to decreases the hydrogen concentration in the anode region and the performance of the fuel cell system decreases. Due to the increased anode pressure can such a performance hit be compensated for at least a period of time. If an error signal signals a missing purge event So by the increased anode pressure a performance slump the fuel cell counteracted.

As already mentioned above, is a dependency the fuel cell voltage to possibly in the region of the cathode incoming hydrogen in all operating situations the at least one fuel cell of the fuel cell system available. Depending on the operating situation may be the collapse of the voltage but comparatively difficult to grasp, because of dynamic Operation and corresponding load jumps anyway different voltage values be detected. For this, if necessary, strongly fluctuating Filter out voltage values to those who for driving and / or monitoring the valve device is to be used, can be comparatively be elaborate.

Therefore It is in a particularly favorable and advantageous development the process of the invention provided that the fuel cell after opening the valve device operated at least cyclically in a region of lower current density becomes. This is a detection of the caused by the hydrogen Voltage drop even easier and more reliable possible.

In an alternative and / or additional embodiment thereof it is envisaged that the opening of the valve device only occurs when a steady state operating condition the fuel cell was detected. Such a stationary Operating state, in particular, an idle operation of the fuel cell be. But also in the regular operation of the fuel cell system arise again and again for a certain period of time stationary operating states of the fuel cell, so even during such stationary operating conditions an opening of the valve device can take place and then a very safe and reliable detection of any eventuality Reaction of the fuel cell voltage to that through the open Valve device in the region of the cathode incoming hydrogen is guaranteed.

There the inventive method, the number of necessary sensors are reduced, and so cost-effective safe and reliable operation of a fuel cell system its preferred application lies in the field of fuel cell systems, as used in means of transport on the Land, be used in the water or in the air. Especially Here is a compact, simple and reliable construction necessary, because the systems compared to stationary systems much higher loads such as vibrations, temperature fluctuations and the like are exposed, and sensors used could fail faster than for example in stationary systems.

Further advantageous embodiments of the invention will become apparent from the remaining dependent claims and are based on the embodiment clearly, which will be described in more detail below with reference to the figures is explained.

there demonstrate:

1 a section of a fuel cell system for operation with the inventive method;

2 a diagram with the time histories of voltage U, hydrogen concentration c (H2) and valve position of a valve for the purge;

3 a possible flowchart for the inventive method for controlling a drain / purge; and

4 a possible flowchart for the inventive method for monitoring the valve device.

In the illustration according to 1 is a fuel cell system 1 in a relevant to the present invention section highly schematized indicated. Most important component of the fuel cell system 1 is a fuel cell 2 which are typically as a stack of single NEN fuel cell, designed as a so-called fuel cell stack. The fuel cell 2 has an anode compartment 3 and a cathode compartment 4 which, in the embodiment shown here, should be separated from one another by a proton-conducting membrane. At the fuel cell 2 So it is a so-called PEM fuel cell stack.

The anode compartment 3 the fuel cell 2 becomes from a hydrogen storage device 5 via a metering valve 6 and a line member with hydrogen from the hydrogen storage device 5 provided. In the area of the anode compartment 3 unreacted hydrogen passes through a recirculation line 7 back to the area where the fresh hydrogen is via the metering valve 6 to the anode compartment 3 flows. This recirculation of the exhaust gas from the anode is total as an anode circuit 8th or anode loop. The recirculation line 7 thus leads in a conventional manner unused gas from the area of the anode compartment 3 back to the anode room 3 wherein the gas is fresh hydrogen from the hydrogen storage device 5 mixed. To the pressure loss in the anode compartment 3 and the recirculation line 7 is in the area of the recirculation line 7 a recirculation conveyor 9 arranged, which for the return of the unused gas from the anode compartment 3 provides. The recirculation conveyor 9 can be designed as a hydrogen circulation blower, as in 1 is indicated. Additionally or alternatively, a gas jet pump would be conceivable, which by the hydrogen from the hydrogen storage device 5 is driven, and the gas from the area of the recirculation line 7 sucks and mixed with the fresh hydrogen and the anode compartment 3 supplies.

The cathode compartment 4 the fuel cell 2 is supplied in the illustrated embodiment with air. The oxygen in the air serves as an oxidant for the chemical reaction inside the fuel cell 2 and forms together with the hydrogen in a conventional manner water, wherein electric power is released, which at the fuel cell 2 can be tapped accordingly. The air for the cathode compartment 4 is via an air conveyor 10 compacted and the cathode compartment 4 fed. For the preparation of the air while other components, such as air filters, intercooler, humidifier or the like may be present, their representation has been omitted here for simplicity. The air conveyor 10 For example, it can be designed as a screw compressor or as a flow compressor. In the area of the recirculation line 7 is also a water separator 11 provided in which during operation of the fuel cell 2 liquid water from the anode circuit 8th collects. In the area of the water separator 11 is to drain this water a valve device 12 in the outlet area of the water separator 11 , so typically in the direction of gravity below, provided. About a line 13 is the valve device 12 with the supply air to the cathode compartment 4 connected.

The fuel cell system 1 also has an electronic control system 14 which is common and common in such systems. Through this control electronics 14 A variety of functions in the fuel cell system are monitored and controlled. Since only two functions are of interest for the present invention, only one connection is the control electronics 14 with the fuel cell 2 indicated, which for measuring or monitoring the voltage U of the fuel cell 2 should serve. Another control line connects the control electronics 14 with the valve device 12 so that the valve device 12 , which is designed for example as a solenoid valve, over part of the control electronics 14 can be controlled accordingly.

This structure corresponds to the corresponding part of a fuel cell system 1 , as is known from the prior art. Due to the fact that the cathode compartment 4 Air is supplied, and that the hydrogen in the anode circuit 8th around the anode compartment 3 is led over time, it comes to an accumulation of nitrogen in the region of the anode circuit 8th since nitrogen passes through the membrane from the air in the cathode compartment 4 in the area of the anode compartment 3 diffused. Nevertheless, a sufficient amount of hydrogen or a sufficiently high concentration of hydrogen in the anode compartment 3 the fuel cell 2 To be able to ensure this nitrogen must be blown off in conventional systems from time to time (purge). Also, the accumulating water must be drained from time to time (drain). In the present embodiment, both the drain and purge are via the water separator 11 and the valve device 12 , As is known from the prior art, alternative structures would also be conceivable, in particular constructions in which a separate valve device is provided in each case for the drain and for the purge. Essentially, the method described below also applies to this, in which case the control of two valve devices instead of the control of the one valve device 12 must be made accordingly.

Now it is like that opening the valve device 12 For example, it can be time-controlled, since based on corresponding empirical values It is known when a correspondingly large amount of nitrogen in the anode cycle 8th has enriched, and when in the water separator 11 so much water has been collected that this must be drained to overflow or in particular a spilling over of the water separator 11 to avoid movements such as occur in a motor vehicle. This would cause water to return to the area of the anode circuit 8th and optionally active surfaces in the region of the anode compartment 3 To block. The opening of the valve device 12 can therefore be timed accordingly. An alternative embodiment may also provide that in the control electronics 14 the amount of accumulated water is estimated by a calculation accordingly. Since it is widely known how much water at which power per unit time in the anode circuit 8th accumulates, can be calculated by simply calculating and adding up the amount of water in each of short periods of a particular power comparatively simple and efficient, the amount of accumulating water. For this purpose, no extremely accurate calculation is necessary because due to the inevitable sloshing always a certain safety distance between the top of the water separator 11 and the maximum water level must be provided. Due to this security, which is necessary in any case, the calculation and accumulation of the accumulating volume of water can be rather rough and thus correspondingly simple.

In addition to the alternative use of a time-controlled opening of the valve device 12 or opening the valve device 12 Based on a calculated amount of water, and in particular when reaching a maximum allowable amount of water, a combination of these two methods can be considered. For example, the amount of water can be calculated accordingly. If, after a certain time has elapsed, a sufficiently large amount has not yet accumulated, this method opens the valve device 12 be triggered, it can be time-triggered forcibly, for example, to dissipate the accumulating nitrogen to a deterioration of the performance of the fuel cell 2 to prevent. At the same time, the water is always removed with it, since both media via the water 11 and the one valve device 12 be dissipated. Likewise, a construction would be conceivable in which this is done the other way around, that is to say the time-controlled opening of the valve device 12 takes place while the control electronics 14 the valve device 12 only then opens on the basis of the calculated water value, if it is to be feared that the water separator 11 could overflow before a time-controlled opening of the valve device 13 he follows.

After the valve device 12 is open, is in the embodiment shown here the 1 first the water over the pipe element 13 drain and with the hot supply air to the cathode compartment 4 mix and typically evaporate in this. Only after the entire amount of water has been drained, is the gas cushion in the water separator 11 above the water a certain amount of gas from the region of the anode circuit 8th over the valve device 12 and the conduit element 13 in the area of the supply air to the cathode compartment 4 stream. This gas will typically contain nitrogen and a small amount of hydrogen.

It has now been found that when hydrogen is introduced, even with already very small amounts of hydrogen, there is a drop in the voltage U across the fuel cell 2 he follows. This dip in the voltage U can now be used to determine if and how much hydrogen over the valve device 12 in the area of the cathode compartment 4 was drained. In particular, the voltage dip can be used to determine that now no more water comes, but a gas, which from the anode circuit 8th and always contains a certain amount of hydrogen. With the collapse of the voltage value, one knows that the water separator 11 is reliably emptied, and that a certain amount of gas from the region of the anode circuit 8th was drained. Accordingly, the detection of the voltage drop across the fuel cell 2 through the control electronics 14 be used to the valve device 12 to drive, so here again close. On a hydrogen concentration sensor, for example in the supply air to the cathode compartment 4 or in the region of the conduit element 13 or in the region of the anode circuit 8th can therefore be waived. Likewise, elaborate and susceptible to water level sensors for detecting the water level in the water 11 be waived. This makes the structure simpler, more compact and less expensive. In addition, the water level sensors in particular are relatively susceptible to contamination and therefore tend to generate faulty signals. By dispensing with such sensors, the fuel cell system 1 So be safer.

In the presentation of the 2 is the course of a voltage U at the fuel cell 2 shown over time t. Among them is the course of the hydrogen concentration c (H2) in the region of the cathode input, which was measured here for experimental purposes by means of an externally connected hydrogen sensor. Underneath is in a third The diagram shows the position of a valve for the Purge, where 1 is the closed position and 0 is the open position. Other than in construction according to 1 For experimental purposes, only a Purge valve was used, which immediately after opening gases, especially hydrogen-containing gases, discharges, since the drained water has no effect on the voltage and the current of the fuel cell 2 has, since the water only moistens the supply air accordingly. The lower diagram now shows at a time t ₀ that the purge valve is opened at this time. Thus, a hydrogen-containing gas reaches the area of the cathode space 4 , This leads, as can be seen in the upper diagram, directly to a collapse of the voltage U at the fuel cell 2 , In the middle diagram, for test and control purposes, the hydrogen concentration C (H2) at the inlet of the cathode compartment 4 measured. It can be seen that the onset of the voltage U is accompanied by an increase in the hydrogen concentration c (H2).

About the voltage dip of the voltage U at the fuel cell 2 So it can be very easily determined that hydrogen-containing gas in the region of the cathode space 4 is present, in the structure shown above so the valve device 12 is open and that in the water separator 11 collected water has already been completely drained. Since the voltage U of the fuel cell 2 in the control electronics 14 detected anyway and used for the purpose of power control, the measurement of the voltage U is no additional effort. Thus, with minimal changes in the software of the control electronics 14 a corresponding signal can be generated as soon as the voltage dip is detected, which is then used to close the valve device 12 can be used again. In addition to the discharge of water and a safe discharge of the nitrogen-containing gas from the anode circuit 8th To realize either a sufficient length of the line must be realized 13 be selected so that a corresponding delay between the measured value of the voltage dip and the outflow of the gas in the region of the valve device 12 is reached. This can be ensured that not only the water is drained, but also a large enough amount of gas from the anode circuit 8th is blown off to reduce the nitrogen concentration accordingly. Alternatively or in addition to a corresponding adjustment of the cable length 13 may also have a time delay between the detection of the voltage drop across the fuel cell 2 and closing the valve device 12 to be built in. In particular, the time delay can be chosen to be as large as the hitherto customary opening time of a timed pure purge valve in a fuel cell system 1 comparable performance.

The experiments of the inventors have shown that the detection of a voltage drop across the fuel cell 2 basically always occurs when hydrogen-containing gas in the area of the cathode compartment 4 arrives. In this case, the relationship is approximately proportional, so that on the basis of the voltage value by which the voltage U breaks down, also on the concentration of hydrogen in the region of the cathode space 4 can be concluded. This too can be used to control the valve device 12 be used according to, is compared to the embodiments set out above, however, a little more complicated, as stored on corresponding maps or the like, this relationship and must be calculated with. However, it is possible, the valve device 12 to close when reaching a certain hydrogen concentration in the region of the cathode space, ie in particular when falling below a predetermined voltage value. This can ensure that 100 percent of hydrogen passes through the conduit element 13 in the area of the cathode compartment 4 arrives, so that in any case all nitrogen from the region of the anode circuit 8th was drained.

Particularly advantageous for detecting a voltage dip is the operation of the fuel cell at low current density. Under low current density in the context of the invention is to be understood as a current density in the order of 10% or less than the maximum current density. The lower the current density, the more obvious the voltage dip can be seen, and accordingly easier to detect. In order to be able to work safely and reliably with a simple structure, it can therefore be provided that the fuel cell after opening the valve device 12 to operate in an operating state with low current density. This allows the fuel cell 2 however, may not provide the required performance for the duration of the drain and subsequent purge. Because in most systems backup batteries or supercaps are used to run parallel to the fuel cell 2 To mitigate power peaks, however, this should not be a problem in conventional systems, since for these short periods of time the power can be taken from this memory element. Instead of operating the fuel cell 2 during the entire opening time of the valve device 12 In the range of low current density, it could also be provided to cyclically approach the area of low current density. However, the cycles must then be selected in any case so that detection of the quite dynamic voltage dip can be made safely.

Alternatively, it would also be possible to use the fuel cell 2 in such a way that it has a corresponding region of lower current density, at which a measurement voltage can then be tapped, the break-in of which takes the place of the voltage drop of the fuel cell. Such a construction could, for example, be realized such that a certain number of individual cells in the fuel cell, which is typically designed as a fuel cell stack 2 are formed via a variation in the region of their gas diffusion electrode so that they are always operated with a correspondingly low current density. By tapping the voltage of these individual "sensor cells" in the fuel cell 2 Then, the above-described method could be realized without having to provide power through corresponding memory elements, while the fuel cell 2 even works in a "measuring mode" with a lower current density.

Alternatively or additionally, it would also be conceivable to modify each of the individual cells in such a way that they have a region of low current density in the region of their air inlet. Then, however, the voltage U would have to be tapped accordingly inside the cell, which is possible in principle, in terms of effort and tightness of the fuel cell 2 However, it is relatively expensive.

Another possibility is to provide a single cell designed as a sensor cell, or optionally also a group of individual cells, which are parallel to the fuel cell 2 is supplied by the same supply air. This parallel fuel cell, which still in the fuel cell stack of the fuel cell 2 integrated with and can be cooled together with this, would then be used as a pure sensor cell, which serves to detect the voltage drop during the influx of the hydrogen-containing purge gas into the cathode compartment of this fuel cell.

Alternatively to the construction with the one "real" fuel cell 2 and a pure sensor cell, the structure can also be formed so that the fuel cell 2 is constructed as a modular fuel cell. The individual modules of the fuel cell 2 can then be flowed parallel to the supply air. If only one of the fuel cells 2 in the operation with low current density changes, can over the other fuel cell 2 a corresponding compensation of the required power to a certain extent, so that only a small amount of power would have to be fed from the memory devices, while the one module of the fuel cell 2 is in operation with low current density to determine the onset of the voltage U.

The individual methods can be combined as desired with each other, with the method in particular the combination of drain and purge according to the embodiment according to 1 can be realized. Alternatively, of course, the construction with separate drain and purge valves is conceivable, wherein the structure for the purge valve is analogous to the above, in which case to a corresponding time delay or a defined length of the line element 13 can be waived. In the case of a pure purge valve, the voltage drop would occur directly or after flowing through the line element 13 occur, so in this case a time delay or a correspondingly long line of the line element 13 would be necessary.

Based on a flowchart in 3 By way of example, a possible embodiment of the method will be explained in more detail, without limiting the individual points of the method to exactly these sequence steps.

The procedure starts at the box marked Start here. Thereafter, values are reset, in particular the values of a computer or summator, which adds up the volume of accumulating water. After that starts together with the fuel cell 2 the calculation of the volume of accumulating water in individual time periods depending on the average power of the fuel cell 2 in this period. The generated values are added up accordingly. Once the accumulated value of the volume has reached a maximum predetermined volume, this loop is exited and the valve device 12 open. Simultaneously or immediately after opening the valve device 12 is due to operation of the fuel cell 2 changed with low current density. At the latest then the measurement of the voltage U of the fuel cell is 2 one. Usually, the measurement of the voltage is carried out continuously anyway, so that here only uses a corresponding monitoring on a voltage dip. If the voltage U has collapsed, this is a sign that hydrogen-containing gas in the area of the cathode space 4 arrived, all the water from the water separator 11 So safely drained. Then it becomes immediately in the regular operation of the fuel cell 2 switched back to normal power density to restore the full available power. Then a time delay is waited until certainly a sufficiently large amount of gas, the valve device 12 has happened so that no nitrogen in the anode cycle 8th remains. The valve device 12 is then closed and the process starts anew by resetting the values used to calculate the volume of water produced.

Alternatively or in addition to the previously described use of the voltage drop around the valve device 12 to control accordingly, can the knowledge of the invention that by the smallest amounts of hydrogen on the cathode side 4 the fuel cell 2 a voltage dip occurs, also use otherwise. Another preferred use may be, for example, in monitoring the functionality of the valve device 12 lie. For this purpose, an opening of the valve device is targeted 12 through the control electronics 14 triggered, then it is expected that after a predetermined time hydrogen in the cathode region 4 arrives, provided the fuel cell 2 is operated and the valve device 12 and the conduit element 13 are consistent. If they are not, then the expected voltage drop will be absent. Thus, therefore, the functionality of the valve device can be 12 as well as the free flowability of the conduit element 13 check accordingly. This may be particularly advantageous if the fuel cell system 1 is started at temperatures below freezing. In the water separator 11 and optionally also in the region of the conduit element 13 remaining water can be frozen under these circumstances. Due to the very simple structure with a waiver of water level sensors in the water separator 11 and hydrogen concentration sensors in the anode region 3 This state can not be detected without further ado. However, can be inferred by the structure described above in the absence of a voltage dip on such a malfunction, so that a faulty valve device 12 or frozen items 11 . 12 or 13 can be assumed.

In detail, this will be described below in principle with reference to another flowchart. The flow chart of the 4 In addition to this detection, it also includes an alternative possibility of detecting the breakdown of the voltage U.

Unlike the previously described operation of the fuel cell 2 with low current density is at the in 4 Example described operation the targeted monitoring of the valve device 12 or also the drain / purge into a stationary operating state of the fuel cell 2 laid. Accordingly, the flowchart of the 4 with checking for a steady state of the fuel cell 2 , Such a check can for example be made so that continuously both the fuel cell 2 required current values as well as those of the fuel cell 2 supplied current values. For each configurable runtime, these values are then monitored for steady state conditions. This can be done for example by a moving averaging. A comparison of the detected values with low-pass filtered values or the like would also be conceivable. If the values do not exceed deviation gradients, which can previously be determined on the basis of suitable measurements, maps or the like, then a stationary state can be assumed. If now for both the default values of the required nominal current of the fuel cell 2 as well as for the detected values of the actual flow from the fuel cell 2 these conditions are met for a steady state operation, can total from a steady state operation of the fuel cell 2 be assumed.

With this option, it is possible to temporarily use sustained stationary operating states, for example the previously described method for monitoring the drain / purge via the voltage drop across the fuel cell 2 to enable.

Thus, both in the known as stationary operation idling operation of the fuel cell system and during normal operation in short-term stationary states can be very easily by hydrogen on the cathode side 4 detect induced voltage dips. This means that the process does not affect the operation of the fuel cell 2 is limited with low current density, as set forth in the embodiment described above.

In the flowchart of the embodiment according to 4 Now, as soon as such a stationary state is present, the valve device 12 opened accordingly so that the drain / purge starts. At the same time, a timer is started for a maximum predetermined time. As soon as the voltage has broken in, the drain / purge has already been completed to such an extent that the valve device can be closed again immediately or after a certain time delay. If the voltage does not break, the time is still monitored. As soon as the maximum time has been reached without the voltage having collapsed, an error can occur, for example, the valve device 12 , a clogged conduit element 13 or the like can be deduced. The system or the control device 14 then generates an error signal. This can, according to the principle of the representation 4 then otherwise used, which is indicated by the X marked box and will be described later in more detail. After the error signal is generated, it makes sense here, too, the valve device 12 close again, for example, after a possible thawing of the line 13 to generate no malfunction. At the same time, a timer can then be started for a new test. In particular, when the valve device 12 or the conduit element 13 only frozen, this makes sense, because after a certain time it can be expected that thaw the elements and then given the desired functionality again. Once the timing is reached, the system restarts, for example, by a steady state of the fuel cell 2 reviewed.

Unlike in the 4 Of course, it would also be conceivable to nest the test on the steady state and the time test until a new test or build parallel to each other. Instead of or in addition to the passage of time for a new examination would be here in the context of 3 described way to detect a possible amount of water conceivable. In addition to testing for a steady state of the fuel cell 2 of course, it would also be conceivable alternatively or additionally, in this type of process management, the fuel cell 2 at least temporarily operate in a region of lower current density to increase the quality of the detection of the voltage dip.

In addition to the pure detection and monitoring of the functionality of the valve device 12 can be a method as shown in the exemplary illustration of 4 is described, of course, also be used to a startup process, in particular a cold start operation of the fuel cell system 1 from temperatures below freezing point. In general, the operation of a fuel cell system after start-up assumes that, for example, due to frozen lines or the like, no drain / purge is still possible. Nevertheless, a hydrogen concentration on the anode side 3 the fuel cell 2 provide an approximately regular operation of the fuel cell 2 allows in these situations with a correspondingly increased anode pressure. However, this leads overall to a deterioration in the performance of the fuel cell, if this takes place over a longer period of time. Also, if a drain / purge occurs at elevated anode pressure, more hydrogen will be from the anode side due to the higher pressure 3 removed, as actually desired. This hydrogen then passes to the cathode side 4 and may not be fully implemented there due to the amount. This leads to corresponding hydrogen emissions and overall too high a consumption of hydrogen in the fuel cell system 1 ,

Now there is the possibility of the fuel cell system 1 to operate from the start with the increased anode pressure. Then start that in the flowchart of 4 described system, it is checked at regular intervals again and again, whether the valve devices 12 and the conduit element 13 and the water separator 11 freely continuous or still frozen. As long as no voltage dip is detected, such a situation exists. It is therefore always when, from the start, no first voltage dip was detected and / or if an error signal is present, an operation of the fuel cell system 1 selected with increased anode pressure. Only after the voltage U has collapsed a first time and thus a full functionality of the drain / purge is ensured, the anode pressure is lowered accordingly and the fuel cell system 1 can switch to regular operation. If problems occur again during operation, then again, according to the 4 described again generates an error signal and the anode pressure can be increased according to this error signal.

Especially in the starting state of a fuel cell system 1 , especially when this is used in a vehicle, can be expected very often with very dynamic operating conditions. Therefore, except the idling is not a stationary operating state of the fuel cell system 2 In these situations, a start / stop system, which is typically present in such a vehicle, must recognize the fuel cell system 1 during stop phases of the vehicle either shuts off or puts into a standby mode, as long as suppressed, until the first time a voltage dip on the cathode side 4 the fuel cell 2 has been detected. In these situations, idling is a particularly suitable and favorable operation for detecting the voltage dip. However, if the system is shut down during idle, the functionality described above will be disturbed. Therefore, in such situations after the start of the fuel cell system 1 suppress the start / stop functionality until it is certain that the valve device 12 and the conduit element 13 as well as the water separator 11 work safely and are free from, for example, ice blockages.

As from the representation of 4 can be seen, the process has the ability to perform automatically after a predetermined time, the test again and again. Only if over a very long period of time only error signals are generated, which can be done for example by counting the individual error signals, then there is a fault, for example, the valve device 12 which can not be remedied by prolonged operation, ie, for example, thawing of frozen blockages. Then, via the generated error signals, for example, a warning message or a maintenance recommendation for the fuel cell system 1 be issued. In addition, the fuel cell system can 1 over a too high number generated error signals are also turned off.

The method allows it with slight modification of the software in the control electronics 14 to operate a conventional fuel cell system easily and efficiently. Due to this operating method, in the fuel cell system 1 in addition to the use of sensors, in particular the use of level sensors in the water separator 11 and hydrogen concentration sensors in the supply air to the cathode compartment 4 and / or in the region of the anode circuit 8th be waived. The structure can therefore be simplified and can ensure the safe and reliable operation of a fuel cell system 1 guarantee. The method can be used in particular in fuel cell systems 1 applied, which provide electrical energy in means of transport on land, in the water or in the air. The electrical energy can be used in particular for propulsion purposes, but also alternatively or additionally for auxiliary equipment, consumer electronics or the like.

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited patent literature

• - DE 69912918 T2 [0003]
• - WO 2008/052578 A1 [0005]
• - DE 69920279 T2 [0007]

Claims (25)

A method for operating a fuel cell system having at least one fuel cell, wherein an anode space of the fuel cell is supplied with hydrogen, wherein unused hydrogen is returned from the area to the anode space via an anode circuit in the area in front of the anode space, wherein in the region of the anode circuit via at least one valve means water and / or gas is discharged from the region of the anode circuit, wherein the discharged water and / or gas is introduced into a volume flow flowing to a cathode space of the fuel cell, and wherein the electrical voltage across the fuel cell is measured at least temporarily, characterized in that the Actuation of the at least one valve device ( 12 ) takes place at least partially as a function of the measured voltage (U). Method according to claim 1, characterized in that the valve device ( 12 ) is closed in dependence of the measured voltage (U). Method according to claim 2, characterized in that the closing of the valve device ( 12 ) as a function of a voltage drop across the fuel cell ( 2 ) he follows. Method according to claim 2 or 3, characterized in that the closing of the valve device ( 12 ) takes place in response to falling below a predetermined voltage value. Method according to claim 3 or 4, characterized in that the closing of the valve device ( 12 ) delayed after the voltage dip or falls below the predetermined voltage value. Method according to one of claims 1 to 5, characterized in that the valve device ( 12 ) is time-controlled. Method according to one of claims 1 to 6, characterized in that the valve device ( 12 ) as a function of a calculated amount of water in the anode circuit ( 8th ) is opened. Method according to one of claims 1 to 7, characterized in that after opening the valve device ( 12 ) for a maximum predetermined time to a voltage dip at the fuel cell ( 2 ) and if this voltage drop across the fuel cell ( 2 ) does not occur, the valve device is closed again. A method according to claim 8, characterized in that a closing of the valve device ( 12 ) generates an error signal without a previous voltage dip. A method according to claim 8 or 9, characterized in that after closing the valve device ( 12 ) without prior voltage dip the valve device ( 12 ) is opened again after a predetermined time. Method according to claim 9 or 10, characterized in that the fuel cell system ( 1 ) during the presence of an increased anode pressure error signal. Method according to one of claims 9 to 11, characterized in that the fuel cell system ( 1 ) is operated after start up to a first detected voltage dip with increased anode pressure. Method according to one of claims 9 to 12, characterized in that a start / stop functionality of the fuel cell system ( 1 ) is suppressed until after the start of the fuel cell system ( 1 ) a first voltage dip has been detected. Method according to one of claims 1 to 13, characterized in that exactly one valve device ( 12 ) is used as a drain valve in a water separator ( 11 ), which in turn in the anode circuit ( 8th ) is arranged. Method according to one of claims 1 to 14, characterized in that the fuel cell ( 2 ) after opening the valve device ( 12 ) is operated at least cyclically in a region of low current density. Method according to one of claims 1 to 14, characterized in that the fuel cell used ( 2 ) is formed so that it has a current density distribution, which is formed so that the current density in the region of the in the cathode space ( 4 ) entering the volume flow of the supply air is always low. Method according to one of claims 1 to 14, characterized in that the volume flow of the supply air to two parallel fuel cells ( 2 ) or two parallel connected sections of a fuel cell ( 2 ), wherein one of the fuel cells or one of the sections is configured or operated in such a way that at least at times there is a low current density. Method according to one of claims 1 to 17, characterized in that the opening of the Valve device ( 12 ) takes place only when a stationary operating state of the fuel cell ( 2 ) was detected. A method according to claim 18, characterized in that for detecting a stationary operating state of the at least one fuel cell ( 2 ) and that of the at least one fuel cell ( 2 ) is continuously detected, wherein the detected values are checked by a calculation for steady state conditions, and wherein a stationary state of the fuel cell ( 2 ) is displayed if a stationary condition has been detected for both values. Method according to claim 19, characterized the calculation comprises a moving averaging. Method according to claim 19 or 20, characterized that the calculation compares detected values with low-pass filtered values includes. Method according to claim 19, 20, 21, characterized that the values are assumed to be stationary, as long as the change gradient is below a predetermined value. Method according to one of claims 1 to 22, characterized by its application for operating a fuel cell system in a means of transport on land, in the water or in the air. A method according to claim 23, characterized in that the fuel cell system ( 1 ) provides electrical energy for the propulsion and / or auxiliary equipment and / or electronic components of the means of transport. Method according to claim 23 or 24, characterized that the means of transport is a motor vehicle.

Images