WO2004107489A1

WO2004107489A1 - Simplified method of monitoring cells in a fuel cell

Info

Publication number: WO2004107489A1
Application number: PCT/FR2004/050206
Authority: WO
Inventors: Thomas Nietsch; Olivier Verdu
Original assignee: Helion
Priority date: 2003-05-27
Filing date: 2004-05-26
Publication date: 2004-12-09
Also published as: FR2855657B1; FR2855657A1

Abstract

The invention relates to a method of monitoring the correct operation of cells in a fuel cell comprising one or more stacks of unit cells which are connected in series to one another, said stacks being interconnected in series and/or in parallel by means of electrical connections. The inventive method comprises the following steps consisting in: regarding each unit cell stack as two parts (part A and part B) which comprise the same number of unit cells and which are disposed symmetrically in the stack in relation to the centre thereof; and, for each stack, measuring the voltages of parts A and B, and comparing said measurements.

Description

METHOD FOR SIMPLIFIED MONITORING OF CELLS IN A FUEL CELL

DESCRIPTION

TECHNICAL AREA

The invention relates to a simplified method for monitoring the cells of a fuel cell, said method being capable of detecting an operational failure of one or more of these cells.

STATE OF THE PRIOR ART

The current economic development of industrialized countries implies an ever growing need for alternative energy sources, the ideal energy source being as far as possible renewable, high efficiency, silent and developing high power densities. Many sectors of activity constitute potential users of such an energy source and applications are envisaged in both the military and civil sectors. Military applications include the propulsion of submarines, mobile electric generators and low power units, replacing batteries. Commercial applications concern the transport sector (cars, buses, trucks, trains), but also stationary applications, for production localized electricity, and possibly the use of co-generated heat. There are other potential applications in the field of portable and miniaturized devices. The fuel cell is a high-performance electrochemical energy conversion device, which directly converts chemical energy from a fuel, in some cases renewable, into electrical energy. The most advanced technology in the field of temperatures below 100 ^C C is that of polymer electrolyte fuel cells.

A fuel cell is made up of an assembly of elementary cells, stacked on top of each other and arranged in sufficient numbers to ensure the electrochemical production of electricity under desired voltage and current conditions. The fuel cell involves, without mixing them, a fuel (for example hydrogen, methanol, ...) and an oxidizer which is generally oxygen taken from the air.

The operating principle of this electrochemical generator is based on the electrochemical synthesis reaction of water. The cell core consists of an anode, supplied with fuel, a cathode, supplied with oxygen, and an electrolyte. At the anode, the hydrogen undergoes catalytic oxidation which releases protons and electrons. The electrons circulate along the external electrical circuit, while the protons are transported in the electrolyte towards the cathode, where they recombine with electrons and oxygen, under the effect of a catalytic reduction, to produce water.

The efficiency of the fuel cell, theoretically close to unity, reaches practical values twice that of thermal engines. In addition, it is silent and not very polluting, even if the fuel is an organic compound. This good yield and this lower pollution are among the reasons for research and development in this field.

One of the essential components of a polymer electrolyte fuel cell is the membrane-electrode assembly (EME) which comprises a proton exchange membrane and two catalytic gas diffusion electrodes, consisting of an active layer and a layer of diffusion. The EME is installed between two plates ensuring the supply of reactive gases, and this system is closed by two current collectors, the assembly constituting a cell of a fuel cell core.

In the case of a PEM (Proton Exchange Membrane) fuel cell, the electrolyte is a proton exchange membrane. This membrane must fulfill three roles: to be impermeable to gas in order to play its role of gas separator, to be electronically insulating in order to avoid any short circuit, and finally, to be a good ionic conductor for ensuring the transport of protons from the anode. towards the cathode. As can be seen in FIG. 1, the elementary cell 1 is composed of two electrodes, an anode 2 and a cathode 3, separated by an electrolyte 4. In the case of a fuel cell with a proton exchange membrane, the assembly 2, 3, and 4 constitutes an EME. The anode 2 and the cathode 3 are respectively supplied with fuel (fuel entering 5 into the cell and fuel leaving 15 from the cell) and by oxidizer (oxidizer entering 6 into the cell and oxidizer leaving 16 of the cell).

In general, the balance reaction taking place in each elementary cell is:

H ₂ + -O ₂ → π ₂ o

At anode 2, after diffusion through the diffusion layer, the hydrogen is catalytically oxidized in the active layer to give protons and electrons. The electrons use an external electrical circuit towards the cathode 3 and supply the load 7, while the electrolyte 4, which is here a proton exchange membrane, ensures the exclusive transport of the protons from the anode 2 to the cathode 3, but also separation of reactant gases. At cathode 3, the oxygen undergoes a catalytic reduction and recombines with the protons and the electrons to give water. The water produced and any inert substances are removed by excess gas flow, either continuously or periodically using purges.

In practice, several tens of elementary cells are necessary to make a fuel cell of average power. These cells elementals are stacked and combined to obtain said average power. The elementary cells are electrically connected in series within the same stack. The stacks can then be electrically associated in series or in parallel so that the voltage and intensity of the fuel cell can be adjusted as required.

An elementary cell is generally characterized by its polarization curve 11 (see FIG. 2), that is to say by the voltage-current pairs. In normal operation of the fuel cell and knowing the current and the operating conditions which are applied to it, one can predict what the voltage of each elementary cell should be by referring to the reference polarization curve. Thus, if there is a malfunction at the level of a cell, such as for example a rupture of the membrane of the cell or a poor gas supply, this results in a reduction of the voltage across the terminals of this cell. This fall is generally several hundred mV.

However, the cells being associated in series within the same stack, the current passing through the stack of cells is the same as the current passing through each cell of the stack. By measuring the current of the stack, we thus know the current delivered by each cell.

On the other hand, the measurement of the tension of the stack does not give detailed information on the tension individual cells and therefore on the possible presence of defects in the cells.

The numerous monitoring systems intended to detect faults in the cells of a fuel cell are currently implemented on the principle of measuring the individual voltage of each cell. For example, for a fuel cell composed of a number N of cells, the N cell voltages are measured and then the N signals originating from the cells are processed by a monitoring system. Knowing that for applications targeting powers greater than one kilowatt the number N can vary from 10 to 100, we realize the difficulties involved in the monitoring systems of the prior art. Indeed, to obtain these powers and still have a system capable of monitoring the fuel cell, this in practice requires wiring the N cells and providing a monitoring system associated with each of these N cells. This way of proceeding is therefore damaging in terms of complexity, volume, reliability and cost of the surveillance system. We see in particular that such a system is not suitable for industrial production.

A simplified version of this monitoring system exists and consists in measuring only the voltage of a group of cells typically consisting of 3 to 5 cells. But for a stack of 100 cells, this still involves wiring 20 to 30 groups of cells, which remains restrictive. Another problem with existing monitoring systems is that they are based on measurements of cell or group of cell voltages. In these systems, faults are detected through a reduction in voltage. The voltage of a cell is not constant. It depends on the current flowing through it and on the operating conditions that are applied to the cell. For example, a cell voltage of 0.5V is characteristic of a fault at low current density (less than 0.5 A / cm ² ), but is normal for high current densities (greater than 1 A / cm ² ). To be effective, cell voltage measurements therefore require a perfect knowledge of the polarization curve and the operating conditions of each cell in order to really know whether a voltage is characteristic of a fault or not.

PRESENTATION OF THE INVENTION The object of the invention is to provide a method for monitoring the state of the elementary cells constituting a fuel cell which would not present the problems of the prior art.

This object and others are achieved, in accordance with the invention, by a process for monitoring the proper functioning of the cells of a fuel cell, the fuel cell consisting of one or more stacks (or "stacks"). ”) Of elementary cells connected in series with each other, said stacks being connected together in series and / or in parallel by electrical connections. The process which is the subject of the invention comprises several stages.

First of all, virtually each stack of elementary cells must be separated into two parts (part A and part B) comprising the same number of elementary cells and being arranged symmetrically in the stack of elementary cells with respect to the middle of said stack . Then, for each of these stacks, the voltage of part A (U _A ) and part B (U _B ) is measured. Finally we compare these two measures.

This gives a monitoring of the symmetry and the state of homogeneity of the parts A and B of the stack. Indeed, in the absence of a faulty cell within parts A and B, the voltage of part A (U _A ) and the voltage of part B (U _B ) are similar. On the other hand, in the presence of a faulty cell within parts A and / or B, the voltage of part A (U _A ) and the voltage of part B (U _B ) are different.

Advantageously, the comparison of the voltage measurements of parts A and B is carried out by subtracting the voltage measurements of parts A and B so as to obtain an absolute value. Advantageously, the comparison of the voltage measurements of parts A and B is then carried out by comparing the absolute value with a determined absolute threshold value. Thus, if the absolute value obtained for one or more of the stacks is greater than the determined absolute threshold value, it can be concluded that said stack or said stacks in question contains one or more faulty cells among the cells of part A and / or of part B. Likewise, if the absolute value obtained for one or more of the stacks is less than the determined absolute threshold value, said stack or said stacks in question contains no faulty cells among cells in Part A and Part B.

Note that if parts A and B constitute the entire stack, then the method according to the invention makes it possible to monitor the symmetry and the state of homogeneity of the entire stack.

One of the advantages of this method resides in the fact that a system is obtained which is simple to implement, which is simply based on two voltage measurements, regardless of the number of cells making up the stack. This process also has the advantage of allowing a simplified interpretation and monitoring, since the absolute value of the difference in voltages Δu = | u _A -U _B | parts A and B consisting only of non-failing cells is almost zero, whatever the current of the stack and the operating conditions applied.

In general, the method according to the invention makes it possible to detect the presence in the stack of one or more pierced membranes. Similarly, the method can also make it possible to monitor the homogeneity of the gas supplies to the elementary cells or to monitor the evacuation of the inert elementary cells. Furthermore, the method makes it possible to distinguish different failures, that is to say that it will be possible to distinguish, for example, a failure due to the presence of a hole in a membrane, from a failure due to a poor supply of gas from one or more cells.

Another advantage of this process lies in the fact that it can make it possible to control a purging device, the function of which is to evacuate the water produced in the stack and the various inert materials.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be better understood and other advantages and features will appear on reading the description which follows, given by way of nonlimiting example, accompanied by the appended drawings among which:

FIG. 1 is a diagram illustrating the operating principle of an elementary PEMFC cell, FIG. 2 represents the polarization curve of an elementary cell,

- Figure 3 is a diagram illustrating the principle of the method according to the invention in which the voltages of parts A and B of the stack are measured.

DETAILED PRESENTATION OF PARTICULAR EMBODIMENTS

FIG. 3 illustrates the principle of the monitoring method according to the invention. In this figure, there is shown a stack formed of N cells. The monitoring process involves separating virtually stacking in two parts (A and B). Parts A and B are made up of the same number of elementary cells (m cells). Parts A and B are arranged symmetrically in the stack of elementary cells making up the stack (the middle of the stack is represented by a broken line). In FIG. 3, it can be seen that there are thus parts C and D (having p cells) and a part E (q cells) on either side of parts A and B. The method then consists in measuring and simply compare the voltages, U _A and U _B , of the two parts A and B. Indeed, in theory and in the case of a perfectly symmetrical voltage profile, we have: Δu = | u _A -U _B | = 0V. However, it turns out that in practice, the tension profile is not perfectly symmetrical. There is generally a difference in voltage Δu <0.05V. As the tension profile always has the same shape whatever the operating conditions and the operating point of the stack, Δu is an almost constant value. Consequently, if one or more cells experience any defect (for example, a rupture of the membrane or a poor gas supply), this will cause the voltage to drop and this will therefore make the stack highly "asymmetrical". For example, if it is found that the value of the voltage difference Δu is high (typically greater than 0.3V), it can be concluded that one or more of the cells of parts A and / or B are faulty. In other words, if we take into account that the probability that a simultaneous defect appears on the two parts A and B of the stack and leads to voltage drops which compensate for each other is negligible, we come to the conclusion that if we detect sudden variations in Δu or too high values, we can conclude that one or more cells of the parts are malfunctioning A and B.

To illustrate these various assertions, we take a fuel cell made up of a single stack of elementary cells comprising 34 cells and we carry out three tests in order to compare the results obtained depending on whether the cells are or are not failing, parts A and B being each consisting of 17 cells.

The first test illustrates the behavior of a fuel cell in normal operation, that is to say that no fault is recorded. This test, to be representative of the operation of a fuel cell, is carried out for a period of approximately two hours and alternates operations in stationary regime (constant current) and operations in dynamic regime (the current varying from 10 to 90 AT) . During this test, the individual monitoring of the cell tensions shows that all the cells have the same level of tension and shows the absence of any defect in the cells examined in parts A and B. It is also noted that Δu = | U _A -U _B | is less than 0.1V during the whole test, including in dynamic regime and if the background noise is omitted from the measurement, this value is stable. The second test is representative of the behavior of a fuel cell when certain of these cells are pierced. In this particular case, we see that the cell voltages are very dispersed and that certain voltages fall below the low threshold acceptable for a cell (from Figure 2, this threshold is located at around 0.4-0.5V) . During this test, Δu is not stable and varies from -2V to 8V. By imposing a threshold on Δu around 0.5V, we are thus able to detect such a fault.

Finally, the third test shows the behavior of a fuel cell in the case where certain cells are poorly supplied with gas. It is noted that the voltages of the cells are very dispersed and that certain voltages fall below the low threshold acceptable for a cell (located at around 0.4-0.5V). During this test, it can be seen that Δu is not stable and varies from -1 to IV. By imposing a limit threshold on Δu around 0.5V, we are able to detect the presence of a malfunction in one or more of the cells.

Claims

1. Method for monitoring the proper functioning of the cells of a fuel cell, the fuel cell consisting of one or more stacks of elementary cells connected in series with each other, said stacks being connected together in series and / or in parallel by electrical connections, said method comprising the steps consisting in:

- in each stack of elementary cells, consider two parts (part A and part B) comprising the same number of elementary cells and being arranged symmetrically in the stack of elementary cells with respect to the middle of said stack,

- for each stack,

- measure the voltages of the two parts A and B, - compare these two measurements.

2. Monitoring method according to claim 1, characterized in that the comparison of the voltage measurements of parts A and B is carried out by subtracting the voltage measurements of parts A and B so as to obtain an absolute value.

3. Monitoring method according to the preceding claim, characterized in that the comparison of the voltage measurements of parts A and B is performed by comparing the absolute value with a determined absolute threshold value.