Methods relating to monitoring fuel cells

Abstract

The invention relates to a method of determining water accumulation in and or removal from a fuel cell, the method comprising circulating fuel gas in the anode side of the fuel cell for producing electric energy in a fuel cell process, providing at least two purge pulses from the fuel circulation, analyzing the composition and/or volume of purged gas of said at least two gas purge pulses for determining the amount of water accumulation in and/or removal from the fuel cell.

Claims

1. A method of determining water accumulation in or removal from a fuel cell, the method comprising: circulating fuel gas in an anode side of the fuel cell for producing electric energy in a fuel cell process, providing at least two purge pulses from the fuel circulation, measuring the volume of purged gas during each of said at least two gas purge pulses, and calculating a ratio of a first gas volume and a second gas volume, the first gas volume being removed from the fuel cell during the first purge pulse and the second gas volume being removed during the second of the at least two gas purge pulses, based on said calculated ratio, determining the amount of water accumulated in the fuel cell, wherein said at least two gas purges are provided within a time period of 5 seconds.

2. The method according to claim 1, further comprising measuring fuel gas concentration in the at least two gas purge pulses as part of the determination of at least one of the amount of accumulated or removed water.

3. The method according to claim 1, further comprising providing the at least two gas purge pulses via a purge valve according to a purge valve opening sequence, the at least two purge pulses in said purge valve opening sequence being triggered by at least one of: a predefined purge valve triggering scheme, an elapsed time, an amount of current produced by the fuel cell, and a predefined increase of pressure drop at the anode side.

4. The method according to claim 1, wherein the at least two purge pulses each have a duration of a fraction of a second.

5. The method according to claim 1, wherein the fuel cell has a dead-end configuration.

6. The method according to claim 1, further comprising deriving a parameter descriptive of system performance or aging using said analysis.

7. The method according to claim 1, wherein a pressure drop in the anode side of the fuel cell is under 10 millibars.

8. The method according to claim 1, further comprising measuring gas volumes purged during the at least two purge pulses and determining said amount of water accumulation or removed at least partly based on said gas volumes.

9. The method according to claim 1, wherein the method is carried out in a fuel cell system of a vehicle.

10. The method according to claim 1, wherein the method is carried out in a stationary fuel cell system.

11. The method according to claim 1, wherein the fuel gas is hydrogen.

12. The method according to claim 1, wherein the fuel cell is a polymer electrolyte fuel cell (PEFC).

13. A method of determining water removed from a fuel cell, the method comprising: circulating fuel gas in an anode side of the fuel cell for producing electric energy in a fuel cell process, providing a first gas purge pulse and a second gas purge pulse from the fuel circulation, analyzing the hydrogen composition of the gas purged during the first gas purge pulse and the second gas purge pulse, and determining the amount of removed water by calculating a ratio of the hydrogen concentration of the first gas purge pulse with respect to the second gas purge pulse, wherein the first and second gas purge pulses are sequential and of the same duration, wherein the duration is less than one second and the time between gas purge pulses is within five seconds.

14. The method of claim 13, wherein the gas purge pulses are performed for diagnostic purposes.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows the principal operation of anode side of the hydrogen fuel cell system in dead-anode mode, with recirculation of anode gas.

(2) FIG. 2 shows as a graph the measurement of hydrogen flow and concentration during a double purge in different conditions (d2003-58=wet, d2003-52=dry)

(3) FIG. 3 illustrates the principal operation of anode side of the hydrogen fuel cell system in dead-anode mode, with recirculation of anode gas and description of gas volumes.

DETAILED DESCRIPTION OF EMBODIMENTS

(4) Multiple purge method for detecting water accumulation in anode gas channels

(5) The water accumulation in anode channels is important to measure to optimize the duration and the frequency of purge during normal operation. In addition, the change (increase) in water accumulation under the same operation conditions can be used as diagnostic tool to estimate aging of the bipolar plate and GDL as aging of these components will increase water accumulation.

(6) The practicality and accuracy of the method in this innovation can evaluated against the differential pressure drop method assuming pressure drop is large enough in the anode side of the stack.

(7) The innovation can also be used for detection of bipolar plate and GDL aging on the cathode side, if fuel cell is arranged so that anode and cathode are interchanged.

(8) However, the aging of cathode side components is easily seen from the pressure drop data. Pressure drop on the cathode side is significantly larger and the gas composition is more constant.

(9) The innovation can be used in all fuel cell cars as well, as well as in some stationary systems.

(10) Hydrogen fuel cell systems for automotive and other applications are expected to operate with dead-end mode and with recirculation of hydrogen gas. FIG. 1 illustrates such system.

(11) During the hydrogen purge, the purge valve is typically opened for a fraction of a second.

(12) According to one embodiment, hydrogen purge is applied, when concentration of inert gases or water on the anode side has been accumulated to the level that the performance of the stack is decreased and should be retained.

(13) According to one embodiment, purge is triggered by a voltage signal or it can be triggered periodically from calculated time and/or from the amount of current produced or due to pressure drop increase from the anode side.

(14) Purge can also be performed for diagnostic purposes, if the processes of the purge provide valuable information for the system operation or for the estimation of system aging.

(15) One of the problems on the PEFC is water accumulation in the gas channels on both anode and cathode.

(16) If there is the same flow resistance in the cells, the volume of the purged gas is practically constant, since flow resistance during the purge determines the purged gas volume. If there is water in the channels, the flow resistance will be higher and the purged gas volume will be smaller. The hydrogen purge removes water from the channels. Therefore, if a second hydrogen purge is performed after the first one the purged gas volume should be larger. If there is no water in the channels, which is removed, the purged gas volumes should be equal, since flow resistance will be the same. The increase of purged volume can therefore be used as diagnostic signal for liquid water in the channels.

(17) The multiple purge method has been verified experimentally. FIG. 2 shows the results of experimental verification. As can be seen in the hydrogen flow rate signal, the flow rates (and purged gas volume) are practically constant, when stack is operating in dry conditions (d2003-52=dry) with minimal amount of water in anode gas channels. It can also be seen that in wet conditions (d2003-58=wet) the peak flow rate and purged gas volume of the second gas purge is over twice the flow rate and volume of the first purge.

(18) The measurement of gas flow rates may not be practical in real fuel cell system. The purged volume is not possible to be measured in normal system as this requires high sampling rate and an expensive hydrogen flow meter.

(19) According to one embodiment, purged volume is measured from the change (increase) in the hydrogen concentration, if the purged gas has constant concentration.

(20) From FIG. 2, the different change in the concentration wet vs. dry condition is clearly seen. In wet condition the increase in the second purge is larger, while in dry conditions the increase in the second is smaller.

(21) In every automotive fuel cell system there will be hydrogen concentration sensor.

(22) Therefore, using double/multiple purge this method can be used to detect the presence of liquid water in the anode gas channels.

(23) If the volume of the anode side is known, also absolute purged volume can be measured from the concentration change. If the purged volume is small enough, then only gas with outlet concentration (see FIG. 1) is purged and gas has constant composition.

Method of Analyzing Fuel Cell System Performance Using Tracer Gas and Tracer Gas Sensor

(24) Herein, the use of tracer gas sensor and tracer gas, preferably carbon dioxide, for the measurement of system performance (purged gas volume, fuel efficiency, system efficiency, or some other performance parameter) during use and/or maintenance of a fuel cell system, is described.

(25) According to one embodiment, the inert gas accumulation rate is used for the measurement of purged gas volume (descriptive of fuel efficiency).

(26) According to one embodiment, carbon dioxide concentration is used for measurement of purged gas volume (fuel efficiency). This may be done in steady state use of the fuel cell system or transient use of the system.

(27) According to one embodiment, carbon dioxide or hydrogen concentration is determined for measurement of purged gas composition. Gas composition data can be used for adjusting purge time (valve open) so the hydrogen consumption can be minimized and purged gas volume can be measured accurately.

(28) In the methods, proposed in this invention, the fuel utilization can be measured on-line by using gas sensors (carbon dioxide or hydrogen). The applicability of the methods during normal operation depends on the hydrogen quality used.

(29) The methods can be used on-line during normal operation or during maintenance. During maintenance hydrogen fuel with higher carbon dioxide content can be used for reaching better measurement accuracy. In particular in the maintenance option, carbon dioxide can also be added in the fuel cell system separately from fuel line.

(30) If carbon dioxide or hydrogen is also measured from exhaust anode gas then it can be verified that purged gas contains only gas between stack exit and purge valve.

(31) In first embodiment of the proposed method (with hydrogen sensor) hydrogen fuel utilization can be monitored by measuring total inert gas accumulation rate and using hydrogen purges using so small purged gas volumes that purged gas constant composition.

(32) In this method, both total inert gas content in the used hydrogen, as well as membrane gas permeability must be measured. This is possible, as shown by Karimaki et al. (2011). However, this method may not be practical in commercial applications, as inert gases are accumulated on the anode side also through the membrane and gas permeability of the membrane is dependent on the operating conditions. Inert gas content in hydrogen fuel may also be different in each filling.

(33) In another embodiment of the proposed method (with carbon dioxide sensor) carbon dioxide content is monitored, instead of hydrogen concentration, using carbon dioxide sensor. If the used hydrogen contains sufficient amount of carbon dioxide (2 ppm) accumulated carbon dioxide is coming from the hydrogen and negligible amount is coming through the membrane from the cathode. Since all accumulated carbon dioxide can be assumed to come with hydrogen, the measurement of both carbon dioxide in hydrogen and hydrogen utilization become much simpler than using hydrogen sensor.

(34) Additional carbon dioxide can also be added in the system either using hydrogen fuel containing tens of ppm carbon dioxide or adding carbon dioxide using another feeding line.

(35) When this innovation is used with hydrogen sensor it does not require any additional instrumentation for the vehicle, as fuel cell systems in vehicles will be equipped by humidity and hydrogen concentration sensors. However, the accuracy is limited and the required measurement time is long.

(36) When carbon dioxide sensor is used, more accurate measurement of fuel utilization can be reached. If there is not sufficient amount of carbon dioxide in hydrogen fuel, then carbon dioxide can be added as tracer gas and measurement of fuel utilization can be performed during vehicle maintenance.

(37) With reference to FIG. 3, there is hydrogen gas with 3 carbon dioxide (and inert gas) concentrations in the system: Fresh hydrogen (c.sub.1), Recirculated/purged hydrogen (c.sub.2) and Inlet hydrogen (c.sub.3). In principle there is gradual a transition from c.sub.3 to c.sub.2 in the cells, but this can be assumed to be step-wise.

(38) Use of Steady-State Level

(39) Measurement of Purged Hydrogen Using Steady-State Information:

(40) When purged gas volume is small it contains only gas with concentration c.sub.2. This means that when CO.sub.2 has reached a steady-state, then molar amount of CO.sub.2 that exists from the volumes V.sub.2 and V.sub.3 must be the same as what is fed during the time with gas volume V.sub.1.

(41) When anode gas is purged, then purged gas volume (V.sub.p) with concentration of c.sub.2(CO.sub.2) is replaced by same gas volume with c.sub.1(CO.sub.2)

(42) Then molar balance for the system is then:
c.sub.1*V.sub.1+c.sub.1*V.sub.p=c.sub.2*V.sub.p
and V.sub.p can be calculated easily:
V.sub.p=c.sub.1*V.sub.1/(c.sub.2c.sub.1)

(43) The key issue when using steady-state approach is reliable measurement of c.sub.2 and determination of steady-state.

(44) Use of Accumulation/Decumulation Rate

(45) When gas volumes of V2 and V3 are known, then accumulation rate of carbon dioxide can be used.

(46) Then molar balance for the system is then:
c.sub.1*V.sub.1+c.sub.1*V.sub.p=c.sub.2*V.sub.p+(c.sub.2*V.sub.2+c.sub.3*V.sub.3)
where (c.sub.2*V.sub.2+c.sub.3*V.sub.3) is accumulation/decumulation of carbon dioxide in anode side of the system.

(47) Calculation of V.sub.p is also then straightforward:
V.sub.p=(c.sub.1*V.sub.1(c.sub.2*V.sub.2+c.sub.3*V.sub.3))/(c.sub.2c.sub.1)

(48) If system is initially filled with some concentration of carbon dioxide and pure hydrogen is used, then calculation becomes even more straightforward:
V.sub.p=(c.sub.2*V.sub.2+c.sub.3*V.sub.3)/c.sub.2
Measurement of Single Purge Volume

(49) The use of accumulation/decumulation rate can also be used for measuring a single purge.

(50) The gas volume of single purge can also be calculated using measured carbon dioxide concentration before and after a single purge. Compared to the use of total inert gas amount the method is much more accurate and can be repeated with different condition by adding carbon dioxide as tracer gas in the system and using equation:
V.sub.p=(c.sub.2*V.sub.2+c.sub.3*V.sub.3)/c.sub.2
Measurement of Purged Gas Composition

(51) By measuring carbon dioxide concentration or hydrogen concentration in anode gas volume before the purge and comparing that to purged gas composition (measured from exhaust hydrogen) it can be verified if purged gas has had constant composition.