Fuel cell system, fuel cell control method and program

Abstract

To provide a fuel cell system using a fuel cell using a solid electrolyte which can easily determine whether the fuel cell is in a flooding or dryout condition, preferably control flooding and dryout depending on the results, reduce the number of magnetic sensors required to measure magnetic flux density, and visualize a current distribution for all points on the cathode or anode.SOLUTION: A fuel cell system uses a magnetic sensor 52 that measures magnetic flux density at four points on or near a reactant gas flow path 15a near a cathode or anode of a fuel cell to predict a current distribution on the cathode or anode from the magnetic flux density measured during operation, determines whether the fuel cell is in a flooding or dryout state on the basis of the results, resolves it when flooding is determined, and controls a fuel cell temperature and a reactant gas flow rate to eliminate it when dryout is determined.SELECTED DRAWING: Figure 10

Description

[Technical Field]

[0001] This invention relates to a fuel cell system, a fuel cell control method, and a program, and is particularly suitable for controlling fuel cells using solid electrolytes, such as polymer electrolyte fuel cells. [Background technology]

[0002] Fuel cells generate electricity through a chemical reaction between hydrogen and oxygen. While there are several types of fuel cells, polymer electrolyte membrane fuel cells (also known as proton exchange membrane fuel cells) are currently the most popular due to their low operating temperature and are being applied to cogeneration systems and hybrid electric vehicles. Understanding the internal state and controlling stable power generation is crucial for long-term operation of polymer electrolyte membrane fuel cells.

[0003] In polymer electrolyte fuel cells, flooding, a state of excess water production resulting from the accumulation of water generated by the chemical reaction between hydrogen and oxygen, is one of the major malfunctions. Flooding in the gas diffusion layer, catalyst layer, and gas channels leads to an uneven current distribution, which not only degrades stack performance but can also cause performance variability between cells. As a result, even under nominal operating conditions equivalent to other fuel cells, it becomes impossible to predict fuel cell performance, leading to decreased reliability and poor reproducibility. Fuel cell performance is particularly affected by this malfunction at high current densities because the water removal rate does not match the generation rate. In constant current mode, if the gas channel is blocked by flooding, the battery voltage drops, and water is removed from the battery by periodic purging. The movement and distribution of water within polymer electrolyte fuel cells can be monitored using technologies such as CCD cameras, neutron imaging, water detection paper, X-rays, and MRI. Flooding alters properties such as pressure, local current distribution, and battery resistance, impacting flooding mitigation strategies, including flow field design, operating condition settings, and membrane / electrode assembly (MEA) design.

[0004] Another major malfunction in polymer electrolyte fuel cells is dry-out, a condition where the membrane dries out. Dry-out affects battery performance because the drying of the proton exchange membrane increases the battery resistance. Operating polymer electrolyte fuel cells at high temperatures increases catalytic activity. Output power and water removal rates increase due to the high flow rates of hydrogen and air. Such operation can lead to dry-out. Therefore, to monitor this malfunction, methods such as cell disassembly, temperature measurement, cell resistance measurement (by current interruption method), cell impedance measurement (by Cole-Cole plot), and MRI measurements are used. These results influence strategies to mitigate dry-out, such as controlling flow rates and temperature during operation.

[0005] Therefore, it is important to prevent the occurrence of major malfunctions in polymer electrolyte fuel cell systems, namely flooding and dryout, which affect the performance of the battery.

[0006] To date, various methods have been employed to prevent the aforementioned malfunctions, but it has been difficult to integrate evaluation systems into actual polymer electrolyte fuel cell (MSF) systems. Therefore, current distribution measurements of MSF stacks have been performed using printed circuit boards (PCBs). While this method allows for accurate measurements, it requires electrical connections, increasing the size of the stack. To reduce the cost and size of MSF stacks, non-destructive measurement methods have been developed. Another method proposed in 2005 uses magnetic sensors (see Non-Patent Literature 1). In this method, the current distribution is derived based on magnetic flux density determined by Ampère's law and the Biot-Savart law, and the current distribution in a single fuel cell was measured using a simple apparatus (see Non-Patent Literature 2). Furthermore, to evaluate local current distributions, a three-dimensional finite element simulation method was developed that transforms the magnetic flux of the environment surrounding the MSF stack (see Non-Patent Literature 3). However, these non-destructive measurement methods are difficult to implement in MSF stacks. Furthermore, previous studies have shown that achieving accuracy comparable to that of printed circuit board methods requires the use of numerous magnetic sensors, in addition to considerable computation time. Therefore, there are many challenges in integrating and controlling these methods into polymer electrolyte fuel cell systems. Recently, a method for reducing the number of magnetic sensors has been developed (see Non-Patent Literature 4). However, this method uses an electrical fuel cell model and is not applicable to fuel cell stacks operated by hydrogen and air.

[0007] The inventors previously developed a measurement method using a magnetic sensor inserted into the cooling port (air cooling port) of a polymer electrolyte fuel cell stack for integration into an actual polymer electrolyte fuel cell system. They then evaluated the Nexa® power module in a steady state under flooding and dryout conditions using a fuel cell stack consisting of 20 fuel cells (see Non-Patent Documents 5, 6, and 7). This method proved effective for comparing the measured magnetic field distribution with the magnetic field distribution obtained by simulation (see Non-Patent Document 8).

[0008] However, the conventional measurement method described above, which uses magnetic sensors inserted into the cooling ports of a polymer electrolyte fuel cell stack to measure magnetic flux density, has the disadvantage of requiring a large number of magnetic sensors. Therefore, a reduction in the number of magnetic sensors is desirable, but no effective method for doing so has been proposed until now.

[0009] Against this backdrop, the inventors proposed a control method for a fuel cell in a polymer electrolyte fuel cell stack with the aim of reducing the number of magnetic sensors (Non-Patent Documents 9, 10). In this control method, magnetic sensors are installed on the cathode of the fuel cell to measure the magnetic flux density near the inlet, near two bends, and near the outlet of an air passage that has a winding pattern from inlet to outlet. First, the battery voltage and magnetic flux density of each fuel cell are measured, and the battery voltage of each fuel cell is compared to the reference voltage V s Determine whether the battery voltage is below V s If it is determined that the current distribution on the cathode is below V, the current distribution on the cathode is predicted using a calculation formula based on the magnetic flux density measured by the magnetic sensor, and the battery voltage is V s If it is determined that the current is not below a certain level, the process returns to the step of measuring the battery voltage and magnetic flux density. Based on the predicted current distribution, it is determined whether each fuel cell is in a flooding state. If it is determined that the fuel cell is in a flooding state, it is purged. Then the process returns to the step of measuring the battery voltage and magnetic flux density. If it is not determined that each fuel cell is in a flooding state, it is determined whether each fuel cell is in a dry-out state. If it is determined that the fuel cell is in a dry-out state, it is cooled by a fan. Then the process returns to the step of measuring the battery voltage and magnetic flux density. If it is not determined that the fuel cell is in a dry-out state, the operation is stopped. [Prior art documents] [Non-patent literature]

[0010] [Non-Patent Document 1] K.-H.H.P.W.Stolten, “Magnetotomography-a new method for analyzing fuel cell performance and quality”,J Power Sources,vol.143,no.67-74,2005

Non-Patent Document 2

Non-Patent Document 3

Non-Patent Document 4

Non-Patent Document 5

Non-Patent Document 6

Non-Patent Document 7

Non-Patent Document 8

Non-Patent Document 9

Non-Patent Document 10

Summary of the Invention

Problems to be Solved by the Invention

[0011] However, although the fuel cell stacks described in Non-Patent Documents 9 and 10 can visualize the current distribution for each of the four divided sections on the cathode, it is actually difficult to visualize the current distribution for all points on the cathode because the computational load is too high.

[0012] Therefore, the problem to be solved by this invention is to be able to easily determine whether a fuel cell is in a flooding or dry-out state, to be able to perform good control of flooding and dry-out according to the determination result, to reduce the number of magnetic sensors required for measuring the magnetic flux density, and moreover, to provide a fuel cell system using a solid polymer fuel cell, a control method for a fuel cell, and a program for this fuel cell control method that can visualize the current distribution for all points on the cathode or anode.

[0013] Since flooding and dry-out are problems that can occur in fuel cells using solid electrolytes in general, the above fuel cell system, fuel cell control method, and program for this fuel cell control method can be applied to fuel cells using solid electrolytes in general.

[0014] Therefore, more generally, the problem to be solved by this invention is to be able to easily determine whether a fuel cell is in a flooding or dry-out state, to be able to perform good control of flooding and dry-out according to the determination result, to reduce the number of magnetic sensors required for measuring the magnetic flux density, and moreover, to provide a fuel cell system using a fuel cell using a solid electrolyte, a control method for a fuel cell, and a program for this fuel cell control method that can visualize the current distribution for all points on the cathode or anode.

Means for Solving the Problem

[0015] In order to solve the above problems, this invention At least one fuel cell using a solid electrolyte, A magnetic sensor that measures the magnetic flux density at at least four points in the flow path through which the reaction gas flows, near the cathode or anode of the fuel cell, or near the flow path, A control device that predicts the current distribution on the cathode or anode from the magnetic flux density at at least four points measured by the magnetic sensor during operation of the fuel cell, determines from the predicted current distribution whether the fuel cell is in a flooding or dry-out state, controls the temperature and / or flow rate of the fuel cell to eliminate the flooding if it is determined that the fuel cell is in a flooding state, and controls the temperature and / or flow rate of the reaction gas to eliminate the dry-out if it is determined that the fuel cell is in a dry-out state, This is a fuel cell system that has [the following features].

[0016] Fuel cells using solid electrolytes include, but are not limited to, polymer electrolyte membrane fuel cells or solid oxide fuel cells. A fuel cell may be a single fuel cell, which is the smallest power generation unit, or a fuel cell stack consisting of multiple fuel cells stacked on top of each other. The reaction gases of polymer electrolyte membrane fuel cells or solid oxide fuel cells, i.e., the fuel (reducing agent) and the oxygen-containing gas (oxidizing agent), can be conventionally known and selected as needed.

[0017] The location and number of points for measuring magnetic flux density in the flow path of the reaction gas, or in the vicinity of the cathode or anode of the fuel cell, are appropriately selected according to the flow path pattern, flow path length, fuel cell size, etc., and the number and installation location of magnetic sensors are determined accordingly. Typically, magnetic sensors are installed so that magnetic flux density can be measured at multiple locations in the flow path or in the vicinity of the flow path along the flow path from inlet to outlet. In this case, the points for measuring magnetic flux density by magnetic sensors typically include at least four points near the inlet of the flow path, near the outlet of the flow path, and near bends in the flow path, if any. When a single fuel cell is used, the magnetic sensors are installed near the anode or cathode of the fuel cell, or in the space between fuel cells (e.g., a cooling port) when a fuel cell stack is used. However, they do not necessarily need to be installed permanently; if necessary, the magnetic sensors may be moved from outside the fuel cell toward the fuel cell and positioned near the anode or cathode of the fuel cell, or inserted into the space between fuel cells. If no space such as a cooling vent is provided between fuel cells, magnetic sensors may be embedded in, for example, a bipolar plate or separator. Typically, the cathode or anode is divided into several regions, each containing a point for measuring magnetic flux density, and the current distribution is predicted for each region, but this is not limited to this. In one typical example, the magnetic flux density is measured at four points near the four corners of the cathode or anode, on or near the flow path through which the reaction gas flows, and typically four magnetic sensors are installed accordingly. In this case, for example, the cathode or anode is divided into four regions, each containing one of the four points for measuring magnetic flux density, and the current distribution is predicted for each region.

[0018] As described above, the cathode or anode is divided into four regions, and the calculation for predicting the current distribution in each region is specifically performed as follows: That is, the direction of the current flowing perpendicular to the plane of the cathode or anode is taken as the z-axis, and the x-axis and y-axis are taken parallel to that plane and perpendicular to each other, and the number of the above regions k is 1 to n (n=4), and the x-component B of the measured magnetic flux density is x and component B in the y direction y In contrast, the current component J in the x direction x and the current component J in the y direction y of

number

number

[0019] If a fuel cell system has a fuel cell stack consisting of N fuel cells (where N is an integer greater than or equal to 2) connected in series, the magnetic sensor is typically inserted between the (N-1) / 2nd fuel cell and the (N+1) / 2nd fuel cell when N is odd, and between the N / 2nd fuel cell and the (N+2) / 2nd fuel cell when N is even, but is not limited to this, and may be inserted in other locations. For example, the magnetic sensor may be inserted between the 1st and 2nd fuel cells, or between the (N-1)th and Nth fuel cells, or located outside the 1st or Nth fuel cell, regardless of whether N is odd or even.

[0020] Furthermore, this invention, The steps include: measuring and recording the magnetic flux density at at least four points in the flow path through which the reaction gas flows, near the cathode or anode of the fuel cell, at a predetermined current value immediately after the start of operation of the fuel cell using a solid electrolyte, using a magnetic sensor; The steps include measuring the magnetic flux density at the above current value during the operation of the above fuel cell, The steps include: predicting the current distribution on the cathode or anode that has changed from the initial state using a calculation formula; The steps include determining whether the fuel cell is in a flooding state based on the predicted current distribution, When it is determined that the fuel cell is in a flooded state, the procedure includes controlling the temperature of the fuel cell and / or the flow rate of the reaction gas to resolve the flooding. After controlling the temperature of the fuel cell and / or the flow rate of the reaction gas to eliminate the flooding, the uniformity of the current distribution is evaluated, and it is determined whether the current distribution is non-uniform based on predetermined criteria. If it is determined to be non-uniform, the temperature of the fuel cell and / or the flow rate of the reaction gas are controlled again. If it is determined not to be non-uniform, the control is stopped and the process returns to the step of measuring the magnetic flux density. If the fuel cell is not determined to be in a flooded state, the step is to determine whether or not the fuel cell is in a dry-out state. When it is determined that the fuel cell is in a dry-out state, the step of controlling the temperature of the fuel cell and / or the flow rate of the reaction gas to resolve the dry-out, After controlling the temperature of the fuel cell and / or the flow rate of the reaction gas to eliminate the dryout described above, the uniformity of the current distribution is evaluated, and it is determined whether the current distribution is non-uniform based on the above criteria. If it is determined to be non-uniform, the temperature of the fuel cell and / or the flow rate of the reaction gas are controlled again. If it is determined not to be non-uniform, the control is stopped and the process returns to the step of measuring the magnetic flux density. This is a method for controlling a fuel cell.

[0021] Furthermore, this invention, This is a program that instructs a computer to execute the fuel cell control method described above.

[0022] In the invention of the fuel cell control method described above and the invention of a program for causing a computer to execute this fuel cell control method, the information described above in relation to the fuel cell system invention is valid, as long as it does not contradict their nature. [Effects of the Invention]

[0023] According to this invention, it is possible to easily determine whether a fuel cell is in a flooding or dry-out state, to effectively control flooding and dry-out according to the determination result, to reduce the number of magnetic sensors required to measure magnetic flux density, and to visualize the current distribution for all points on the cathode or anode. [Brief explanation of the drawing]

[0024] [Figure 1] This figure shows a block diagram of a fuel cell system according to one embodiment of the present invention. [Figure 2] This is a perspective view showing the fuel cell stack of a fuel cell system according to one embodiment of the present invention. [Figure 3] This is an exploded perspective view showing a fuel cell constituting a fuel cell stack in a fuel cell system according to one embodiment of this invention. [Figure 4] This is a schematic diagram showing a fuel cell stack of a fuel cell system according to one embodiment of this invention, where the number of fuel cells is two. [Figure 5] This is a schematic diagram showing the case where the number of fuel cells in the fuel cell stack of a fuel cell system according to one embodiment of this invention is N. [Figure 6] This is a plan view showing an example of a gas flow path pattern provided in the separator or bipolar plate of a fuel cell stack in a fuel cell system according to one embodiment of the present invention. [Figure 7] This is a flowchart showing a method for controlling a fuel cell in a fuel cell system according to one embodiment of this invention. [Figure 8] This is a schematic diagram showing the configuration of a fuel cell system according to an embodiment. [Figure 9] This is a plan view showing an example of a magnetic sensor used to measure magnetic flux density in a fuel cell system according to an embodiment. [Figure 10] This is a schematic diagram showing the installation positions of magnetic sensors on the cathodes of each fuel cell in the fuel cell stack in the fuel cell system according to the embodiment, along with the gas flow path through which air is passed. [Figure 11] This is a schematic diagram showing the current distribution on the cathode of each fuel cell in the fuel cell stack before and after purging in the fuel cell system according to the embodiment. [Figure 12] This is a schematic diagram showing the current distribution on the cathode of each fuel cell in the fuel cell stack before and after cooling during operation in the fuel cell system according to the embodiment. [Modes for carrying out the invention]

[0025] The embodiments for carrying out the invention (hereinafter referred to as "embodiments") will be described below with reference to the drawings.

[0026] <One Embodiment> [Fuel cell system] Figure 1 shows a fuel cell system according to one embodiment. As shown in Figure 1, the fuel cell system consists of a fuel cell stack 100 in which multiple fuel cells using solid electrolytes are stacked and connected in series, and a control device 200 that controls the operation of each fuel cell in the fuel cell stack 100. The fuel cells are polymer electrolyte fuel cells, solid oxide fuel cells, etc. The control device 200 includes a computer and memory, and the computer has a program installed that is created according to the flowchart shown in Figure 7, which will be described later, and controls the operation of the fuel cells.

[0027] Figure 2 shows the configuration of the fuel cell stack 100. As shown in Figure 2, in the fuel cell stack 100, multiple fuel cells 10 are stacked and connected in series, and current collector plates 20 and 30 are provided at both ends.

[0028] Figure 3 shows the configuration of one fuel cell 10. As shown in Figure 3, the fuel cell 10 has separators 14 and 15 on both sides of a MEA (Meteorological Energy Acquisition) which integrates an anode 12 and a cathode 13 via a catalyst layer and a gas diffusion layer (not shown) on both sides of a solid electrolyte membrane 11. Separator 14 is provided with a gas channel through which a fuel selected according to the type of fuel cell 10 flows. Separator 15 is provided with a gas channel through which a gas containing at least oxygen, such as air or oxygen, flows. For example, if fuel cell 10 is a polymer electrolyte fuel cell, hydrogen or alcohol is used as fuel. If fuel cell 10 is a solid oxide fuel cell, hydrogen, carbon monoxide, hydrocarbons, etc. are used as fuel.

[0029] Figure 4 shows a fuel cell stack 100 consisting of two fuel cells 10 as an example. This fuel cell stack 100 is an air-cooled type. The solid electrolyte membrane 11 of the fuel cell 10 is not shown. Here, it is assumed that the fuel cell 10 is a polymer electrolyte fuel cell, but the same is basically true if the fuel cell 10 is a solid oxide fuel cell. As shown in Figure 4, in this fuel cell stack 100, a cooling port 40 is provided between the two fuel cells 10, and a magnetic sensor 52 mounted on the tip of a probe 51 is inserted into this cooling port 40. In Figure 4, only one magnetic sensor 52 is shown, but in reality, at least four, and more magnetic sensors 52 as needed, are used. The magnetic sensor 52 may be permanently fixed inside the cooling port 40, or it may normally be outside the fuel cell stack 100 and inserted into the cooling port 40 when measurement is performed. Hydrogen (H2) and air, which are used as fuel, are introduced into the hydrogen inlet 21 and air inlet 22 provided on the current collector plate 20, respectively. The hydrogen and air introduced through the hydrogen inlet 21 and air inlet 22 are then supplied to each fuel cell 10 of the fuel cell stack 100. The hydrogen and air discharged through each fuel cell 10 are then discharged through the hydrogen outlet 31 and air outlet 32, respectively, provided on the current collector plate 30.

[0030] Figure 5 shows a fuel cell stack 100 consisting of N fuel cells 10 (where N is an integer greater than or equal to 3). This fuel cell stack 100 is of the air-cooled type. The solid electrolyte membrane 11 of the fuel cell 10 is not shown in the illustration. The fuel cells 10 constituting the fuel cell stack 100 are numbered from 1 to N. The magnetic sensor 52 is inserted between the (N-1) / 2nd fuel cell 10 and the (N+1) / 2nd fuel cell 10 when N is odd, and between the N / 2nd fuel cell 10 and the (N+2) / 2nd fuel cell 10 when N is even. Accordingly, the cooling port 40 is provided between the (N-1) / 2nd fuel cell 10 and the (N+1) / 2nd fuel cell 10 when N is odd, and between the N / 2nd fuel cell 10 and the (N+2) / 2nd fuel cell 10 when N is even. Figure 5 shows a case where N is an even number, a cooling port 40 is provided between the N / 2th fuel cell 10 and the (N+2) / 2nd fuel cell 10, and a magnetic sensor 52 is inserted into this cooling port 40. The magnetic sensor 52 may be inserted between the 1st fuel cell 10 and the 2nd fuel cell 10, as shown by the dashed line in Figure 5, or between the (N-1)th fuel cell 10 and the Nth fuel cell 10. Furthermore, the magnetic sensor 52 may be located outside the 1st fuel cell 10 or outside the Nth fuel cell 10.

[0031] The gas flow path patterns provided in separators 14 and 15 are selected as needed. Figures 6A and 6B show examples of gas flow path 15a of separator 15. The gas flow path 15a shown in Figure 6A has a meandering pattern that curves left and right from the air (or oxygen) inlet to the outlet. The gas flow path 15a shown in Figure 6B has a pattern in which multiple gas flow paths 15a extending in a straight line from the inlet to the outlet are provided parallel to each other.

[0032] [How to use a fuel cell system] Figure 7 is a flowchart showing the control method for the fuel cell 10 in this fuel cell system.

[0033] As shown in FIG. 7, in step S1, the operation of each fuel cell 10 constituting the fuel cell stack 100 is started.

[0034] In step S2, immediately after the start of operation of each fuel cell 10, the magnetic flux density at a predetermined current value nA is measured and recorded in the memory. The magnetic flux density is measured by the magnetic sensor 52 inserted into the cooling port 40. Step S2 needs to be executed because it is necessary to measure the magnetic flux density immediately after the start of operation, that is, the initial magnetic flux density, which is the magnetic flux density when the fuel cell 10 is in a sound state. The x-component (B 1x, … , B 4x ) of the magnetic flux density in Equation (3) and the y-component (B 1y, … , B 4y ) in Equation (4) correspond to this.

[0035] In step S3, the magnetic flux density at the current value nA is measured during the operation of each fuel cell 10. Step S3 is executed because it is necessary to calculate the variation from the initial state in step S4 described later.

[0036] In step S4, the current distribution on the cathode 13 or anode 14 that has changed from the initial state is predicted by a calculation formula using the magnetic flux density measured by the magnetic sensor 52.

[0037] In step S5, it is determined whether or not each fuel cell 10 is in a flooding state based on the current distribution predicted in step S4.

[0038] If it is determined in step S5 that the fuel cell 10 is in a flooding state, in step S6 the control device 200 controls the temperature and / or the flow rate of the reaction gas of the fuel cell 10 to resolve the flooding. Then, in step S7 the uniformity of the current distribution is evaluated and it is determined whether or not the current distribution is non-uniform based on predetermined criteria. If it is determined that the current distribution is non-uniform, the process returns to step S6 and the control device 200 controls the temperature and / or the flow rate of the reaction gas of the fuel cell 10 again. If it is determined that the current distribution is not non-uniform, in other words, uniform, the control by the control device 200 is stopped in step S8 and the process returns to step S3.

[0039] If it is not determined in step S5 that each fuel cell 10 is in a flooding state, then in step S9, it is determined whether or not each fuel cell 10 is in a dry-out state.

[0040] If it is determined in step S9 that the fuel cell 10 is in a dry-out state, in step S10 the control device 200 controls the temperature and / or the flow rate of the reaction gas of the fuel cell 10 to resolve the dry-out. Then, in step S11 the uniformity of the current distribution is evaluated and it is determined whether or not the current distribution is non-uniform based on the above criteria. If it is determined that the current distribution is non-uniform, the process returns to step S10 and the control device 200 controls the temperature and / or the flow rate of the reaction gas of the fuel cell 10 again. If it is determined that the current distribution is not non-uniform, in other words, uniform, the control by the control device 200 is stopped in step S8 and the process returns to step S3.

[0041] If it is not determined in step S9 that the fuel cell 10 is in a dry-out state, then in step S12, the operation of each fuel cell 10 is stopped.

[0042] [Examples] (Fuel cell system) We prototyped a fuel cell system and conducted evaluation experiments by actually operating it.

[0043] This fuel cell system is shown in Figure 8. As shown in Figure 8, in this fuel cell system, the fuel cell stack 100 is housed inside a rectangular housing 300. An air-cooled 50W class solid polymer fuel cell stack is used as the fuel cell stack 100. The number of fuel cells 10 constituting the fuel cell stack 100 is 5. A hydrogen inlet 21 and an air inlet 22 are provided on one side of the housing 300, and a hydrogen outlet 31 and an air outlet 32 ​​are provided on the side opposite to this side. The coordinate system is set as shown in Figure 8. In this case, the width direction of each fuel cell 10 is the x-axis direction, the height direction of each fuel cell 10 is the y-axis direction, and the direction perpendicular to the surface of the fuel cell 10 (the stacking direction of the fuel cells 10) is the z-axis direction. The five fuel cells 10 constituting the fuel cell stack 100 are numbered 1, 2, ..., 5, starting from the side of the housing 300 where the hydrogen inlet 21 and air inlet 22 are provided and moving toward the side opposite this side. Hereinafter, these first to fifth fuel cells 10 will be referred to as battery 1, battery 2, ..., battery 5, as needed. The hydrogen inlet 21 and air inlet 22 are located on the side of the first fuel cell 10 of the fuel cell stack 100, and the hydrogen outlet 31 and air outlet 32 ​​are located on the side of the fifth fuel cell 10 of the fuel cell stack 100. A hydrogen cylinder 62 is connected to the hydrogen inlet 21 via piping 61. A mass flow meter 63 is installed in piping 61 between the hydrogen inlet 21 and the hydrogen cylinder 62. By opening the mouth of the hydrogen cylinder 62, hydrogen is supplied to the hydrogen inlet 21 via piping 61 while the flow rate is controlled by the mass flow meter 63. An air pump 65 is connected to the air inlet 22 via piping 64. A mass flow meter 66 is installed in piping 64 between the air inlet 22 and the air pump 65. Air is drawn in by operating the air pump 65, and the air is sent to the air inlet 22 via the piping 64 while the flow rate is controlled by the mass flow meter 66. Non-humidified hydrogen was used as the hydrogen source, and non-humidified air was used as the air source. A Takatsuki:YR-30 was used as the air pump 65, and Azbil Corporation's CMS0050 was used as the mass flow meter 63 and 66.The temperature of the fuel cell 10 in the fuel cell stack 100 can be controlled using a cooling fan (not shown). In addition, purging is employed as a flood control method, and water generated in the fuel cell 10 is discharged by opening the hydrogen outlet 31 and increasing the gas flow rate.

[0044] The bottom of the housing 300 is open, and a magnetic sensor 52 mounted on the tip of a probe 51 is inserted vertically into the cooling port 40 from below the fuel cell stack 100 inside the housing 300 (in Figure 8, the magnetic sensor 52 is shown shifted vertically downwards). The detection signal (magnetic flux density) from the magnetic sensor 52 is sent to the I2C interface 72 via the I2C signal line 71, and this I2C interface 72 is sent to a personal computer (PC) 74 connected by a USB cable 73 for processing. A program created according to the flowchart shown in Figure 7 is installed on the personal computer (PC) 74. Figure 9 shows the probe 51 and the magnetic sensor 52 mounted on its tip. As shown in Figure 9, the base of the probe 51 is attached to a circuit board 53, and the wiring of the magnetic sensor 52, which runs through the inside of the probe 51, is connected to the circuit of the circuit board 53, allowing the detection signal to be taken out to the outside via a connector mounted on the circuit board 53.

[0045] A voltage logger (digital logger) 80 is connected to the enclosure 300 for measuring the battery voltage of each fuel cell 10. A commercially available GRAPHTEC:GL240 was used as the voltage logger 80. A 160W DC load 90 is also connected to the enclosure 300.

[0046] (Operating conditions for fuel cell systems) Table 1 summarizes the operating conditions for each malfunction state. Here, the fuel cell stack was operated while reproducing two malfunctions: flooding and dryout. Flooding was reproduced by closing the hydrogen outlet 31 and operating at a low temperature (45±5°C). When a voltage drop occurred under flooding conditions, a 20-second purge was performed. Dryout was reproduced by opening the hydrogen outlet 31 and operating at a high temperature (50~80°C). When a voltage drop occurred in the fuel cell 10 during a dryout state, temperature control was performed using a cooling fan incorporated into the fuel cell stack 100. Reference voltage V s The voltage was set to 0.3V. Flooding and dryout control methods are shown in Table 1. This fuel cell system was operated for 1 hour in constant current mode at 15A.

[0047] [Table 1]

[0048] (Measurement system) The magnetic flux density was measured by inserting a magnetic sensor 52 from below into the cooling port 40 provided between each fuel cell 10 of the fuel cell stack 100, moving upward. In the coordinate system set up as shown in Figure 8, the components of the magnetic flux density in each axial direction are B x ,B y ,B z The output is as follows. The magnetic sensor 52 was inserted into the cooling port 40 between the second fuel cell 10 and the third fuel cell 10 at four points (points 1 to 4) shown in Figure 10. The gas flow path 15a for supplying air (oxygen) to the separator 15 on the cathode 13 side was the one shown in Figure 6A. Because the gas flow path 15a is winding, the magnetic sensor 52 was installed near the inlet (point 1), near the two bends (points 2 and 3), and near the outlet (point 4) of the gas flow path 15a. The dimensions of each part are as shown in Figure 10. The magnetic flux density was measured at 0.5-second intervals using these four magnetic sensors 52.

[0049] (Current distribution calculation) As already mentioned, the magnetic flux density is measured by inserting the magnetic sensor 52 into the cooling port 40. According to the Biot-Savart law, the magnetic flux density produced by the electric current is expressed as follows:

number

[0050] Since the direction of the current vector in the fuel cell 10 is constant, by this equation (1), the vector (r s The direction of vector (r) is determined by the positional relationship between the measurement point and the calculation point. s If the direction determined by (-r) is known, the increase or decrease in magnetic flux density can be determined.

[0051] When the fuel cell 10 is operated at a constant current, even if a current imbalance occurs inside the fuel cell 10, the total current J in the healthy current distribution at the start of operation and the total current J' in the current distribution when a malfunction occurs (total current at the time of calculation) are

number

number

number

number

[0052] In addition to the current intensity at these four points, a boundary condition of current value = 0A, i.e., current intensity = 0, is set, assuming that virtually no chemical reactions occur at the four vertices of the rectangular current distribution surface (the four points indicated by black circles in Figure 10). From the current intensity at these eight points, the current intensity distribution (power generation intensity distribution) corresponding to the current distribution is estimated. The change in the current intensity distribution from the start of operation at each measurement point is calculated. Specifically, the current distribution surface of the fuel cell is divided into n sections, and the current intensity is allocated to the n sections from the above eight points. This allocated current intensity maintains the magnetic flux density distribution when power generation starts or when the fuel cell 10 is in a healthy state, and the difference in current intensity distribution from the healthy state is shown for each measurement point.

[0053] (Results of an experiment to determine the current intensity distribution on the cathode of fuel cell 10) The fuel cell stack was operated in a flooded state, and when the battery voltage fell below 0.3V, the flooding state was resolved by purging for 20 seconds, after which the system was operated in a dry-out state.

[0054] Figures 11A-D show the current intensity distribution before and after purging during the transition from a flooding state to a dry-out state. Figure 11A shows the current intensity distribution before purging (750 seconds after the start of operation). The upper left and lower right of Figure 11A correspond to the inlet and outlet of the air channel 15a shown in Figure 10. The current intensity distribution on the inlet side of the air channel 15a is higher compared to the initial state. In contrast, the current intensity distribution in the area from the lower left to the upper right of the center of Figure 11A is lower compared to the initial state. Subsequently, purging was performed between 760 and 780 seconds. As a result, the battery voltage recovered to above 0.3V. Figure 11B shows the current intensity distribution after 800 seconds. From Figure 11B, the current intensity distribution on the inlet side of the air channel 15a remains high, and the current intensity distribution in the area from the lower left to the upper right of the center remains low. This indicates that water was effectively removed from the air channel 15a by purging. However, it is thought that a small amount of water remains in the MEA of some fuel cells. At this point, the overall current intensity remains low. Figures 11C and 11D show the current intensity distribution after 850 seconds and 900 seconds, respectively. From Figures 11C and 11D, it can be seen that the current intensity distribution gradually becomes more uniform over time.

[0055] Figures 12A-D show the current intensity distribution before and after cooling by the cooling fan during the transition from a dry-out state to a flooding state. Figure 12A shows the current intensity distribution before cooling (after 1050 seconds). From Figure 12A, it can be seen that before cooling, the current intensity distribution is low on the inlet side of the air passage 15a and high on the outlet side. This is consistent with the theoretical consideration that the current intensity distribution decreases on the inlet side of the air passage 15a due to dry air and increases on the outlet side due to the increase in moisture from the generated water. Cooling was performed during the operating period from 1063 seconds to 1195 seconds. Figures 12B and C show the current intensity distribution after 1100 seconds and 1150 seconds, respectively. As can be seen from Figures 12B and C, the current intensity distribution does not change even as time passes. Figure 12D shows the current intensity distribution after 1200 seconds. From Figure 12D, it can be seen that the current intensity distribution is uniform, similar to when purging is performed in a flooding state.

[0056] (Conclusion) Based on the results above, the method of the embodiment allows for the visualization of the fuel cell's current intensity distribution at all points on the cathode, thereby enabling accurate identification of non-uniformity in the current intensity distribution caused by flooding or dry-out conditions, and also enabling accurate identification of the elimination of non-uniformity after purging or cooling. Furthermore, by operating the fuel cell with a uniform current intensity distribution, the performance of the fuel cell can be maintained for a long period. In addition, since the control shown in Figure 7 can be stopped and returned to normal control once the non-uniformity in the current intensity distribution is eliminated, energy is saved.

[0057] As described above, according to this embodiment, the magnetic flux density during operation of the fuel cell 10 is measured by a magnetic sensor 52 inserted into the cooling port 40 between the fuel cells 10 of the fuel cell stack 100, the current distribution on the cathode 13 is predicted from the magnetic flux density at at least four points measured by the magnetic sensor 52 during operation of the fuel cell 10, and it is determined from the predicted current distribution whether the fuel cell 10 is in a flooding or dry-out state. If it is determined that the fuel cell 10 is in a flooding state, the flooding is resolved, and if it is determined that the fuel cell 10 is in a dry-out state, the temperature and / or flow rate of the reaction gas of the fuel cell 10 are controlled by the control device 200 to resolve the dry-out. Therefore, it is possible to easily determine whether the fuel cell 10 is in a flooding or dry-out state, and flooding and dry-out can be controlled effectively according to the determination result, and the number of magnetic sensors 52 required to measure the magnetic flux density can be reduced. In addition, the current component J in the x direction is determined based on equations (3) and (4). x and the current component J in the y direction yBy calculating this, the computational load can be kept low, and the current intensity distribution at each point on the cathode or anode of the fuel cell 10 can be visualized, as shown in Figures 11A-D and 12A-D. Since control can be performed based on this visualized current intensity distribution, the fuel cell 10 can be operated with a highly uniform current intensity distribution, and consequently, the lifespan of the fuel cell 10 can be improved.

[0058] Although embodiments and examples of this invention have been described in detail above, this invention is not limited to the embodiments and examples described above, and various modifications based on the technical idea of ​​this invention are possible.

[0059] For example, the numerical values, shapes, structures, and arrangements mentioned in the above embodiments and examples are merely examples, and different numerical values, shapes, structures, and arrangements may be used as needed. [Explanation of Symbols]

[0060] 10…Fuel cell, 11…Electrolyte membrane, 12…Anode, 13…Cathode, 14, 15…Separator, 20…Current collector plate, 21…Hydrogen inlet, 22…Air inlet, 30…Current collector plate, 31…Hydrogen inlet, 32…Air inlet, 40…Cooling port, 51…Probe, 52…Magnetic sensor, 100…Fuel cell stack, 200…Control device

Claims

1. At least one fuel cell using a solid electrolyte, A magnetic sensor that measures the magnetic flux density at at least four points in the flow path through which the reaction gas flows, near the cathode or anode of the fuel cell, A control device that predicts the current distribution on the cathode or anode from the magnetic flux density at at least four points measured by the magnetic sensor during operation of the fuel cell, determines from the predicted current distribution whether the fuel cell is in a flooding or dry-out state, controls the temperature and / or flow rate of the reaction gas of the fuel cell to eliminate the flooding when it is determined that the fuel cell is in a flooding state, and to eliminate the dry-out when it is determined that the fuel cell is in a dry-out state, A fuel cell system having

2. The fuel cell system according to claim 1, wherein the points on which the magnetic flux density is measured by the magnetic sensor include at least four points near the inlet of the flow path, near the outlet of the flow path, and near the bend in the flow path if there is a bend in the flow path.

3. The fuel cell system according to claim 1, wherein the magnetic sensor measures the magnetic flux density at four points in the vicinity of the four corners of the cathode or anode and in the flow path.

4. The fuel cell system according to claim 3, wherein the cathode or anode is divided into four regions, each containing one of the four points for measuring magnetic flux density, and the current distribution is predicted for each region.

5. When the direction of the current flowing perpendicular to the cathode or anode is defined as the z-axis, and the x-axis and y-axis are taken perpendicular to each other in a direction parallel to that plane, and the section numbers k are 1 to n (n=4), the x-direction component B of the measured magnetic flux density is x and component B in the y direction y In contrast, the current component J in the x direction x and the current component J in the y direction y of [Math 3] [Math 4] The fuel cell system according to claim 4, wherein the current distribution is determined by calculation using the above method.

6. The fuel cell stack consists of N fuel cells (where N is an integer of 2 or more) connected in series with each other. The fuel cell system according to claim 1, wherein the magnetic sensor is inserted between the (N-1) / 2 fuel cell and the (N+1) / 2 fuel cell when N is odd, and between the N / 2 fuel cell and the (N+2) / 2 fuel cell when N is even.

7. The fuel cell system according to claim 1, wherein the fuel cell is a polymer electrolyte fuel cell or a solid oxide fuel cell.

8. The steps include: measuring and recording the magnetic flux density at at least four points in the flow path through which the reaction gas flows, near the cathode or anode of the fuel cell, at a predetermined current value immediately after the start of operation of the fuel cell using a solid electrolyte, using a magnetic sensor; The steps include measuring the magnetic flux density at the above current value during the operation of the above fuel cell, The steps include: predicting the current distribution on the cathode or anode that has changed from the initial state using a calculation formula; The steps include determining whether the fuel cell is in a flooding state based on the predicted current distribution, When it is determined that the fuel cell is in a flooded state, the procedure includes controlling the temperature of the fuel cell and / or the flow rate of the reaction gas to resolve the flooding. After controlling the temperature of the fuel cell and / or the flow rate of the reaction gas to eliminate the flooding, the uniformity of the current distribution is evaluated, and it is determined whether the current distribution is non-uniform based on predetermined criteria. If it is determined to be non-uniform, the temperature of the fuel cell and / or the flow rate of the reaction gas are controlled again. If it is determined not to be non-uniform, the control is stopped and the process returns to the step of measuring the magnetic flux density. If the fuel cell is not determined to be in a flooded state, the step is to determine whether or not the fuel cell is in a dry-out state. When it is determined that the fuel cell is in a dry-out state, the step of controlling the temperature of the fuel cell and / or the flow rate of the reaction gas to resolve the dry-out, After controlling the temperature of the fuel cell and / or the flow rate of the reaction gas to eliminate the dryout described above, the uniformity of the current distribution is evaluated, and it is determined whether the current distribution is non-uniform based on the above criteria. If it is determined to be non-uniform, the temperature of the fuel cell and / or the flow rate of the reaction gas are controlled again, and if it is determined not to be non-uniform, the control is stopped and the process returns to the step of measuring the magnetic flux density. A method for controlling a fuel cell.

9. A program for causing a computer to execute the fuel cell control method described in claim 8.

Images