JP2011028937A - Fuel cell system and operation method of fuel cell system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a fuel cell system for humidifying anode gas. <P>SOLUTION: The fuel cell system includes a cell stack (2) formed by laminating a plurality of unit fuel cells constituted by sandwiching an electrolyte membrane by an anode and a cathode. The fuel cell system carries out operation for supplying fuel gas to the anode so as to repeat a unit cycle where the unit cycle is a cycle of pressure pulsation including a process for increasing the pressure of an anode gas flow passage inside the cell stack and a process for reducing the pressure. The fuel cell system includes a determination means (51) for determining whether the electrolyte membrane is in a dry state or a moist state during operation, and a pressure increase rate changing means (51) for increasing a rate of a pressure increase per unit cycle from a determination result when the electrolyte membrane is in the dry state as compared with in the moist state of the electrolyte membrane. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a fuel cell system and a method for operating the fuel cell system, and more particularly to a method for humidifying a reaction gas.

  As a means for humidifying the reaction gas, a site (humidification area) consisting only of an electrolyte layer having no catalyst layer is provided outside the reaction area (active area) of the MEA, and water at the outlet of one gas flow path is passed through the electrolyte layer. There is a proposal of a so-called cell internal humidification method in which the reaction gas is humidified by being moved to the other gas channel inlet (see Patent Document 1).

JP 2008-97891 A

  However, according to the technique of the above-mentioned Patent Document 1, the relative humidity on the outlet side of the anode gas is extremely low during low load operation, so even if a humidification area is provided outside the active area of the MEA, Since the humidifying capacity is reduced, the anode gas cannot be humidified.

  Therefore, an object of the present invention is to provide a fuel cell system capable of humidifying an anode gas.

  The fuel cell system of the present invention includes a cell stack in which a plurality of unit fuel cells each having an electrolyte membrane sandwiched between an anode and a cathode are stacked, and a process in which the pressure in the anode gas flow path inside the cell stack is increased and the pressure is reduced. The operation of supplying the fuel gas to the anode is performed so that the unit period is repeated with the period of the pressure pulsation consisting of the above process as a unit period. During the operation, it is determined whether the electrolyte membrane is in a dry state or a wet state, and when the electrolyte membrane is in a dry state based on the determination result, the rate of pressure increase per unit cycle is set in the wet state of the electrolyte membrane. Make it bigger.

  According to the present invention, when the electrolyte membrane is in a dry state, the time during which the anode gas flow rate (forward direction) is zero or more (existing) in the process of increasing the pressure of the anode gas flow path is the wet state of the electrolyte membrane. Since it becomes longer than a certain time, the amount of movement of water from the upstream side to the downstream side of the anode gas channel increases. Therefore, humidification of the anode gas channel can be promoted.

It is a schematic block diagram of the fuel cell system of this embodiment. It is a schematic block diagram of a cell stack. It is a top view of MEA of a prior art. It is a top view of MEA of a prior art. It is the sectional view on the AA line of FIG. It is a timing chart which shows each change of the operating state of the hydrogen pressure regulation valve per unit cycle, anode gas channel pressure, and anode gas flow in the anode dead end operation of the prior art. It is a characteristic view which shows the relationship between an operation parameter and a dry condition evaluation parameter | index. It is a characteristic view which shows the relationship between a dry condition evaluation index | exponent and the ratio of the pressure | voltage rise per unit period. It is a timing chart which shows each change of the operating state of the hydrogen pressure regulation valve in the unit cycle, anode gas channel pressure, and anode gas flow in the anode dead end operation of a 1st embodiment. It is a timing chart which shows each change of the operating state of the hydrogen pressure regulation valve in the unit cycle, anode gas channel pressure, and anode gas flow in the anode dead end operation of a 2nd embodiment. It is a schematic block diagram of the fuel cell system of 3rd Embodiment. It is a flowchart for demonstrating opening / closing control of the drain valve of 3rd Embodiment. It is a timing chart which shows each change of the operating state of the hydrogen pressure regulation valve in the unit cycle, anode gas channel pressure, and anode gas flow in the anode dead end operation of a 4th embodiment. It is a timing chart which shows each change of the operating state of the hydrogen pressure regulation valve in the unit cycle, anode gas channel pressure, and anode gas flow in the anode dead end operation of a 5th embodiment. It is a timing chart which shows each change of the operating state of the hydrogen pressure regulation valve in the unit cycle, anode gas channel pressure, and anode gas flow in the anode dead end operation of other examples of a 1st embodiment.

  FIG. 1 is a schematic configuration diagram of a fuel cell system 1 of the present embodiment. The fuel cell system of this embodiment uses a polymer electrolyte fuel cell that is relatively small and excellent in power generation efficiency, and is mounted on a vehicle.

  The cell stack 2 is supplied with a reaction gas (fuel gas and oxidant gas) used for an electrochemical reaction and a cooling medium for cooling the cell stack 2. Hydrogen is supplied to the anode of the cell stack 2 from a hydrogen tank 3 storing high-pressure hydrogen through a fuel gas supply pipe 4. Instead of the hydrogen tank 3, hydrogen may be generated by a reforming reaction using alcohol, hydrocarbon or the like as a raw material. The fuel gas supply pipe 4 is provided with a pressure regulating valve 5 (fuel gas flow rate adjusting means, upper limit pressure adjusting means) for adjusting the supply amount of hydrogen. The cell stack 2 is connected to one end of a discharge pipe 6 for discharging impurities (product water, nitrogen, etc.) together with the fuel gas not consumed at the anode to the outside of the cell stack 2.

  A water separator tank 7 (gas storage means) is connected to the other end of the discharge pipe 6, and the water separator tank 7 stores water vapor in the fuel gas as condensed water. A pipe 8 for discharging the accumulated water is provided below the water separator tank 7, and a normally closed drain valve 9 (volume adjustment mechanism) is provided on the pipe 8.

  In order to discharge nitrogen contained in the fuel gas after the water vapor is separated in the water separator tank 7, a pipe 10 is provided at the upper part of the water separator tank 7, and a nitrogen purge valve 11 (exhaust valve) is provided in the pipe 10. ing.

  FIG. 2 is a schematic configuration diagram of the cell stack 2. The cell stack 2 is configured by stacking a plurality of unit fuel battery cells (single cells) 41. The single cell 41 has a membrane electrode assembly (hereinafter referred to as “MEA”) in the center of the laminated structure. The MEA 42 has a structure in which an electrode catalyst layer and a gas diffusion layer are sequentially laminated on both surfaces of an electrolyte membrane. One side of the electrolyte membrane is used as a cathode, and the other side is used as an anode. A cathode side separator 43 and an anode side separator 44 made of carbon or metal, which are conductive members, are arranged on both surfaces of the MEA 42. An air (oxidant gas) flow path 45 is formed on the surface of the cathode side separator 43 facing the MEA 42, and a cooling water flow path 47 is formed on the opposite surface. A flow path 46 of hydrogen (fuel gas) is formed on the surface of the anode separator 44 facing the MEA 42, and a cooling water flow path 47 is provided on the opposite surface. After stacking a plurality of single cells 41 formed in this way, manifolds 49 and 50 for distributing air, hydrogen and cooling water to each single cell 41 are provided at both ends. 2, air, hydrogen, and cooling water supplied from outside are distributed to each single cell 41. In addition, in order to efficiently perform water circulation inside the cell stack 2, the air flow path 45 and the hydrogen flow path 46 are made to face each other.

  Hereinafter, air supplied to the cathode is also referred to as “cathode gas”, and hydrogen supplied to the anode is also referred to as “anode gas”. The air channel 45 is also referred to as a “cathode gas channel”, and the hydrogen channel 46 is also referred to as an “anode gas channel”.

  Air is supplied from the compressor 15 through the supply pipe 16 to the cathode of the cell stack 2. Instead of the compressor, air supply means such as a blower can be used. The air discharged from the cathode of the cell stack 2 is released into the atmosphere via the discharge pipe 17. A pressure regulating valve 18 is disposed in the discharge pipe 17 to adjust the back pressure (pressure in the cathode gas flow path 45).

  Cooling water is further supplied to the cell stack 2 from the radiator 21 via the pipe 23. Instead of the cooling water, an antifreeze such as ethylene glycol or a cooling medium such as air can be used. Cooling water whose temperature has risen due to the heat generated in the cell stack 2 is sent to the radiator 21 via the pipe 22 and cooled, and then is circulated inside the cell stack 2 again. A circulation pump 24 for water circulation is disposed in the pipe 23. A three-way valve 25 is provided in the pipe 22.

  The fuel cell system 1 configured as described above is a device that directly converts chemical energy of fuel into electrical energy, and supplies an anode gas to the anode and a cathode gas to the cathode. Electric energy is extracted from the electrode by utilizing the following electrochemical reaction that occurs on the surface on the electrolyte membrane side.

Anode reaction: H ₂ → 2H ⁺ + 2e ⁻ (a)
Cathode reaction: 2H ⁺ + 2e ⁻ + (1/2) O ₂ → H ₂ O (b)

  When hydrogen is supplied to the anode, the reaction formula (a) proceeds at the anode and hydrogen ions are generated. The generated hydrogen ions permeate (diffuse) through the electrolyte membrane in a hydrated state to reach the cathode. When oxygen is supplied to the cathode, the reaction formula (b) proceeds at the cathode. As the electrode reactions of these formulas (a) and (b) proceed at each electrode, the cell stack 2 generates an electromotive force. In the formula (b), moisture generated at the cathode is discharged to the outside of the cell stack 2 as water vapor or liquid water by the cathode gas, or passes through the electrolyte membrane and discharged to the outside of the cell stack 2 by the anode gas. The

  Since the gas discharged to the discharge pipe 6 includes the anode gas that has not been consumed for power generation at the anode, the anode gas is wasted if it is discarded as it is. For this reason, fuel gas is supplied to the anode so that the unit period is repeated, with the period of pressure pulsation consisting of the process of increasing and decreasing the pressure of the anode gas flow path 46 in the cell stack 2 as a unit period. Thus, an anode dead end operation is performed in which the fuel gas supplied to the anode is not discharged outside the cell stack 2 and the water separator tank 7.

  The anode dead end operation will be further described. In FIG. 1, the hydrogen pressure regulating valve 5 is opened to supply fuel gas to the anode inside the cell stack 2, and the compressor 15 is started and air is pumped (supplied) to the cathode inside the cell stack 2 to generate power with the MEA 42. Start. When the MEA 42 starts power generation, water is generated at the cathode along with power generation. The generated water moves from the cathode toward the anode and reaches the anode. The water (brine water / liquid water) that has passed through the anode reaction surface 48 (see FIG. 3) permeates the gas diffusion layer in the anode and exits onto the anode gas flow path 46. If the power generation is continued as it is, the pressure of the anode gas flow path 46 remains stuck to the upper limit pressure determined by the hydrogen pressure regulating valve 5, and the fuel gas supplied from the tank 3 is a mass consumed by the power generation. Only flow rate.

  When the cell stack 2 is operated by flowing the mass flow rate, a dynamic pressure sufficient to drain the water on the anode gas channel 46 to the water separator tank 7 cannot be obtained. It has been found from experiments by the inventors that the fuel gas diffusion is inhibited to cause a voltage drop due to insufficient supply of the fuel gas, and the MEA 42 becomes unable to generate power.

  In order to avoid this problem, if the hydrogen pressure regulating valve 5 is temporarily fully closed during power generation, the fuel gas is not supplied from the tank 3 to the cell stack 2, and the drain pressure valve 5 is discharged from the hydrogen pressure regulating valve 5. Power generation is continued using the fuel gas remaining in the anode gas flow path up to 11. In this case, the water separator tank 7 provided in the discharge pipe 6 has the largest volume, and the fuel gas remaining in the water separator tank 7 flows into the anode gas flow path 46 inside the cell stack 2. It flows toward. Then, the gas pressure in the anode gas flow path 46 and the water separator tank 7 inside the cell stack 2 is decreased in order to consume the volume of the anode gas flow path 46 and the fuel gas remaining in the water separator tank 7 by power generation. come.

  When the gas pressure decreases, the hydrogen pressure regulating valve 5 is opened again. Then, the fuel gas from the tank 3 flows toward the anode flow path 46 inside the cell stack 2, and the pressure of the anode gas flow path 46 inside the cell stack 2 increases. The dynamic pressure generated at this time causes the water on the anode gas flow path 46 inside the cell stack 2 to move from the downstream side of the anode gas flow path 46 to the water separator tank 7, so that power generation can be continued to some extent. That is, the process of increasing and decreasing the pressure of the anode gas flow path 46 inside the cell stack 2 is repeated at a constant cycle. As described above, the fuel gas is supplied to the anode so that the unit period is repeated with the period of the pressure pulsation including the process of increasing and decreasing the pressure of the anode gas flow path 46 in the cell stack 2 as the unit period. The operation, that is, the operation in which the fuel gas supplied to the anode is not discharged to the outside of the cell stack 2 and the water separator tank 7 is called the anode dead end operation.

  In the anode dead end operation, the flow direction of the fuel gas is switched. Therefore, when the anode gas flow path 46 in the cell stack 2 is pressurized, the flow direction from the hydrogen pressure regulating valve 5 toward the anode in the cell stack 2 is the forward direction. The gas flow flowing in the forward direction is defined as “forward flow”. Further, the flow direction from the water separator tank 7 toward the anode gas flow path 46 in the cell stack 2 when the anode gas flow path 46 in the cell stack 2 is depressurized is reversed, and the gas flow flowing in the reverse direction is “back flow”. Defined in

  During the operation of the cell stack 2, water generated at the cathode is also supplied to the anode through the electrolyte membrane. In many cases, the electrolyte membrane has high gas permeability. When air is used as the cathode gas, nitrogen that has permeated from the cathode to the anode is deposited on the anode gas flow path 46. Therefore, in the anode dead end operation, the liquid water and nitrogen accumulated on the anode gas flow path 46 are removed using the flow (forward flow) of the anode gas during the pressure increase, and discharged from the nitrogen purge valve 11 and the drain valve 9 to the outside. To do.

  The controller 51 is configured as a logic circuit centered on a microcomputer, and more specifically, a CPU (not shown) that executes predetermined calculations in accordance with a preset control program, and various arithmetic processes are executed by the CPU. A ROM (not shown) in which control programs, control data, and the like necessary for storage are stored in advance, and a RAM (not shown) in which various data necessary for various arithmetic processes performed by the CPU are temporarily read and written. And an input / output port (not shown) for inputting and outputting various signals. The controller 51 controls the pressure regulating valves 5 and 18 by driving the compressor 15 so as to obtain a power generation amount corresponding to the load based on the accelerator opening (corresponding to the load) detected by the accelerator opening sensor 52. Then, the MEA 42 in the cell stack 2 is caused to generate power, and the anode dead end operation is performed in which the fuel gas supplied to the anode in the cell stack 2 is not discharged to the outside of the cell stack 2 and the water separator tank 7. The anode dead end operation itself is known (see Japanese Patent Application Laid-Open No. 2007-149630).

  In such a fuel cell system 1, in order to draw out the performance of the electrolyte membrane in the MEA 42 and improve the power generation efficiency, it is necessary to keep the moisture state of the electrolyte membrane optimal. For this reason, the anode gas and the cathode gas introduced into the cell stack 2 are humidified.

  In this case, if a device for humidifying the cell stack 2 is provided outside the cell stack 2, the system may be complicated. Further, when the fuel cell system 1 is mounted on the vehicle, the system 1 needs to be simplified in order to reduce the size of the system, and humidification circulation can be performed inside the cell stack 2 without using a humidifying device outside the cell stack 2. Ideal.

  Therefore, in order to promote humidification circulation inside the cell stack 2, a counter flow structure is generally used in which the anode gas and the cathode gas are flowed opposite to each other with the electrolyte membrane interposed therebetween. That is, as shown in FIGS. 2 and 3, the anode gas upstream side and the cathode gas downstream side are opposed to the anode gas downstream side and the cathode gas upstream side across the electrolyte membrane. . Here, FIG. 3 is a plan view of the MEA 42 of the prior art (see Japanese Patent Application Laid-Open No. 2008-97891) showing a cell internal humidification method. In the prior art, as shown in FIG. 3, a portion (humidification area 49) consisting only of an electrolyte layer having no catalyst layer is provided outside the active area 48 (anode reaction surface) of the MEA 42 as a means for humidifying the reaction gas. ing. That is, even if the cathode gas supplied to the cell stack 2 is dry, the cathode gas is wetted by the water generated by the cathode reaction on the downstream side of the cathode gas channel 45. The electrolyte membrane has a property of permeating water vapor, and the driving force is determined by the difference in water vapor concentration between the downstream side of the cathode gas channel 45 wetted by the generated water and the upstream side of the dried anode gas channel 46. The water vapor is transported from the cathode gas side to the anode gas side. The anode gas humidified on the upstream side transports water vapor to the cathode gas channel 45 side on the downstream side of the anode gas channel 46, this time opposite to the above. In the case of the counter flow structure, the humidification device outside the cell stack 2 can be eliminated by performing the humidification circulation inside the cell stack 2 in this way.

  However, in the low load operation, the relative humidity on the outlet side of the cell stack 2 in the cathode gas channel 45 is extremely low, so that the downstream side of the anode gas channel 46 cannot be humidified. That is, since the flow rate of the anode gas is small and the amount of water vapor stored in the gas is small, the relative humidity on the outlet side of the cell stack 2 in the anode gas flow path 46 is very likely to decrease. For this reason, even if the humidification area 49 is provided outside the active area 48 of the MEA 42, if the humidification on the downstream side of the anode gas channel 46 is insufficient, the upstream side of the cathode gas channel 45 cannot be humidified. This will be described with reference to FIG. 4. FIG. 4 is a plan view of the MEA 42 of the prior art again. On the low load side, the electrolyte membrane in the region 1 upstream of the cathode gas flow channel 45 in FIG. 4 is first dried, and power generation is not performed. When power generation is not performed in region 1, power generation corresponding to the load is performed in regions 2 to 5, but after a while, the electrolyte membrane in region 2 is dried and power generation is performed in region 2. No longer done. When power generation is not performed in the region 2, power generation corresponding to the load is performed in the regions 3 to 5, but after a while, the electrolyte membrane in the region 3 is dried and power generation is performed in the region 3. No longer done. Such a phenomenon occurs in a chain, and depending on conditions, finally, only the region 5 is in a state of generating power, and the voltage of the entire cell stack 2 is significantly reduced.

  Further details will be described with reference to FIG. FIG. 5 is a model diagram (a cross-sectional view taken along line AA in FIG. 4) of a single cell used in the cell stack 2 in FIG. FIG. 5A shows operating conditions on the high load side where the flow rate of the fuel gas supplied to the anode (the fuel gas supplied to the anode is hereinafter referred to as “supply fuel gas”) is large, and FIG. It shows what happens to the behavior of water on each of the cathode gas channel 45 and the anode gas channel 46 under the operating conditions on the low load side where the flow rate of the fuel gas is small. Under the operating condition on the high load side where the flow rate of the supplied fuel gas is large, the relative humidity increases on the downstream side of the cathode gas flow path 45 with power generation, and the relative humidity difference between the cathode gas flow path 45 and the anode gas flow path 46 is determined as the driving force. As shown in FIG. 5A, water is back-diffused in the electrolyte membrane of the MEA 42 toward the anode gas channel 46 side, and the upstream side of the anode gas channel 46 is humidified. The water vapor that has flowed out to the anode gas channel 46 is transported to the downstream side of the anode gas channel 46 and humidifies the electrolyte membrane upstream of the cathode gas channel 45 (region 1 in FIG. 4). The problem of drying out does not occur. As described above, a technique for humidifying each other's poles by the counter flow of the anode gas and the cathode gas has been known for a long time. However, as shown in FIG. 5B, the cell stack 2 in the anode gas flow path 46 is small because the amount of water that can be held by the anode gas is small under the low-load operation conditions where the amount of generated water and the flow rate of the supplied fuel gas are small. The outlet side remains dry, and therefore, the electrolyte membrane in the region 1 to which only dry gas (dry gas) is supplied upstream of the cathode gas flow path 45 is dried, and the problem described above with reference to FIG. 4 occurs. It is.

  Here, when thinking again, in the anode dead end operation, the flow (forward flow) of the anode gas from the upstream side of the anode gas channel 46 to the downstream side of the anode gas channel 46 occurs only at the time of pressure increase, By this forward flow, it becomes possible to carry water vapor from the upstream side to the downstream side of the anode gas flow path 46 and humidify the entire anode reaction surface 48. Therefore, in order to prevent the electrolyte membrane from drying on the outlet side of the cell stack 2 in the anode gas flow path 46, it is desirable that the rate of pressure increase per unit cycle is large.

  However, in the prior art (Japanese Patent Laid-Open No. 2007-149630), as shown in FIG. 6, liquid water on the anode gas flow path 46 or from the cathode gas flow path 45 side to the anode gas flow path 46 side. For the purpose of removing permeated nitrogen, the pressure increase time T3 is set shorter than the pressure decrease time T2, and the pressure of the anode gas flow path 46 is rapidly increased. In general, nitrogen that permeates from the cathode gas channel 45 side to the anode gas channel 46 side is forced to flow downstream of the anode gas channel 46. In this case, in the prior art, the pressure increase time T3 (the ratio of pressure increase per unit cycle) is not set by distinguishing between when the electrolyte membrane is in a dry state and when the electrolyte membrane is in a wet state. Even when the electrolyte membrane on the cell stack 2 outlet side of the anode gas flow path 46 (cell stack 2 inlet side of the cathode gas flow path 45) is in a dry state, as in the operating conditions on the low load side, the anode gas If the time for generating the anode gas flow (forward flow) from the upstream side to the downstream side of the flow path 46 is short, water vapor cannot be sufficiently conveyed from the upstream side to the downstream side of the anode gas flow path 46, and the anode Drying of the electrolyte membrane tends to occur on the cell stack 2 outlet side of the gas flow path 46 (the cell stack 2 inlet side of the cathode gas flow path 45).

  Therefore, in this embodiment, when the anode dead end operation is performed, when the electrolyte membrane is in a wet state as in the operation condition on the high load side, and when the electrolyte membrane is in the dry state as in the operation condition on the low load side. And set the rate of boosting per unit cycle separately. That is, during the anode dead end operation, it is determined whether the electrolyte membrane is in a dry state or a wet state. From this determination result, when the electrolyte membrane is in a dry state, the electrolyte membrane determines the rate of pressure increase per unit cycle. Larger than when wet.

  According to this embodiment, when the electrolyte membrane is in a dry state, the time during which the anode gas flow velocity (forward direction) is increased to zero or more (exists) in the process of increasing the pressure of the anode gas flow path 46 (pressure increase time T3). However, since the electrolyte membrane is longer than when the electrolyte membrane is in a wet state, the amount of movement of water from the upstream side to the downstream side of the anode gas channel 46 increases. In this way, by increasing the amount of water movement from the upstream side to the downstream side of the anode gas flow channel 46, the cell stack 2 outlet side of the anode gas flow channel 46, and hence the cell stack 2 inlet of the cathode gas flow channel 45. Humidification on the side can be promoted.

The unit period in the anode dead end operation is the sum of the pressure increase time T3 that is a time required for the pressure increase process and the pressure decrease time T2 that is a time required for the pressure decrease process. The “ratio of” is a value defined by a value obtained by dividing the pressure increase time T3 by the sum of the pressure increase time T3 and the pressure decrease time T2. If this is expressed by an expression,
Ratio of pressure increase per unit cycle = T3 / (T2 + T3) (1)
It becomes the following formula.

In the above-described anode dead end operation, the fuel gas is supplied to the anode so as to repeat the process of increasing and decreasing the pressure of the anode gas flow path 46 in the cell stack 2. The target anode dead end operation is not limited to that described above. For example, as shown in FIG. 6, it is also an object to provide a process for maintaining the pressure of the anode gas flow path constant between the pressure increasing process and the pressure decreasing process. At this time, the unit period in the anode dead end operation is a maintenance time T1 required for the process of maintaining the pressure constant, a pressure reduction time T2 which is a time required for the pressure reduction process, and a time required for the pressure increase process. Since it is the sum of the boosting time T3, the “ratio of boosting per unit period” is
Ratio of pressure increase per unit cycle = T3 / (T1 + T2 + T3) (2)
The value defined by the formula

  In the following, the case (2) above, that is, the case where the unit period in the anode dead end operation is the sum of the maintenance time T1, the pressure reduction time T2, and the pressure increase time T3 will be mainly described.

To determine whether the electrolyte membrane is dry or wet,
<1> Temperature of the cell stack 2,
<2> When a cooling medium for cooling the cell stack 2 is provided, the temperature of the cooling medium on either the upstream side or the downstream side of the cell stack 2;
<3> The flow rate of the cathode gas supplied to the cathode (the cathode gas supplied to the cathode is hereinafter referred to as “supply cathode gas”),
<4> Supply cathode gas pressure,
<5> Outside air pressure,
<6> Elevation where cell stack 2 is located,
<7> The amount of generated current of the cell stack 2,
<8> Impedance of cell stack 2,
<9> Total voltage of cell stack 2,
<10> At least one piece of information (operation parameter) is used among the voltages of each unit cell. Then, a dry state evaluation index Cdry is calculated based on the operating parameters, and the ratio of the pressure increase time per unit cycle is determined based on the calculated dry state evaluation index Cdry.

  First, the operation parameters will be described in detail. As a means for determining whether the electrolyte membrane is in a dry state or a wet state, whether the electrolyte membrane is in a dry state or a wet state by diagnosing whether or not the power generation state of the cell stack 2 is directly detected. In addition to the means for determining whether or not the electrolyte membrane is in a dry state or a wet state from the operating environment of the cell stack 2, a method of predicting (determining) whether the electrolyte membrane is in a dry state or a wet state can be taken. Here, the operating environment that can be used to predict whether the electrolyte membrane is in a dry state or a wet state is the cell stack temperature of <1> and the cell stack 2 of the cooling medium of <2>. <3> supply cathode gas flow rate, <4> supply cathode gas pressure, <5> external pressure, <6> elevation, <7> above The amount of generated current.

  Here, it is characteristic that the determination of whether the electrolyte membrane is in a dry state or a wet state can also be made from the amount of generated current during anode dead end operation (<7> above). In general, since the electrolyte membrane has high water vapor permeability, there are cases where the water vapor transmission amount is locally higher than the amount of water produced by the cathode reaction accompanying power generation. Since the amount of generated water is particularly small during low-load operation, the water vapor transmission amount is relatively increased, and gas drying on the outlet side of the cell stack 2 of the anode gas flow path 46 tends to proceed easily. Therefore, it is desirable to increase the rate of pressure increase per unit cycle as the operation becomes lighter.

  Whether the electrolyte membrane is in a dry state or a wet state can also be determined from the operating environments of <1> to <6> above. As the volume flow rate of the supply cathode gas being dried (<3> above) is increased, the electrolyte membrane is dried on the outlet side of the cell stack 2 of the anode gas passage 46 (on the inlet side of the cell stack 2 of the cathode gas passage 45). In order to proceed, it can be determined from the flow rate of the supply cathode gas whether it is in a dry state or a wet state. Further, the lower the supply cathode gas pressure (above <4>), the more the supply cathode gas volume flow rate increases and the relative humidity is less likely to rise. In this case, the electrolyte membrane on the cell stack 2 outlet side of the anode gas flow channel 46 (the cell stack 2 inlet side of the cathode gas flow channel 45) is easily dried. Therefore, you may determine whether it is in a dry state or a wet state from supply cathode gas pressure or external pressure. Further, as an alternative to the external air pressure, whether the electrolyte membrane is in a dry state or a wet state may be determined from the altitude (<6> above) where the cell stack 2 is located. In addition, since the saturated vapor pressure of the anode gas increases and the relative humidity decreases as the temperature increases, the higher the temperature of the cell stack 2 (<1>), the higher the temperature of the cell stack 2 (the cathode gas outlet 46 side) (cathode gas). It may be considered that the electrolyte membrane at the cell stack 2 inlet side of the flow path 45 is in a dry state. When the cell stack 2 has means for cooling with the cooling medium, at least one temperature (above <2>) of the cooling medium on the upstream side or the downstream side of the cell stack 2 is replaced with the temperature of the cell stack 2. As described above, it can be determined whether the electrolyte membrane is in a dry state or a wet state.

  Of course, the power generation state of the cell stack 2 may be directly measured to determine whether the electrolyte membrane is in a dry state or a wet state. When the electrolyte membrane is in a dry state, power generation is biased to a high-humidity region in the cell stack 2, so that a voltage drop is likely to occur. Therefore, whether the electrolyte membrane is in a dry state or a wet state can be determined from a decrease in the total voltage (<9>) of the cell stack 2. In addition, the voltage of each unit cell (above <10>) or the voltage between one or more stacked cells is measured by a cell voltage monitor, and from the decrease, whether the electrolyte membrane is in a dry state or a wet state is determined. You may judge.

  In addition, since the resistance of the electrolyte membrane increases when the electrolyte membrane is in a dry state, it can be determined from the impedance (<8>) of the cell stack 2 whether the electrolyte membrane is in a dry state or a wet state. . Here, the impedance of the cell stack 2 is the ratio of the voltage response of the cell stack 2 to the fluctuation of the current applied to the cell stack 2. A typical method for detecting the impedance of the cell stack 2 is to measure the resistance of the cell stack 2 with an AC ohmmeter. Further, instead of using an ohmmeter, it may be determined whether the electrolyte membrane is in a dry state or a wet state from a voltage response to current fluctuation during anode dead end operation. For example, it is possible to superimpose a small alternating current component on the power generation command current amount and to estimate the resistance of the electrolyte membrane from the voltage response. In addition, a case where a large voltage drop occurs quickly with respect to a transient load change may be determined as a time when the electrolyte membrane is in a dry state.

  FIG. 7 shows the relationship between the above operating parameters and the dry state evaluation index Cdry, and FIG. 8 shows the relationship between the dry state evaluation index Cdry and T3 / (T1 + T2 + T3), which is the rate of pressure increase per unit cycle. .

  As shown in the above <1> to <10>, the unit of the operation parameter is various. Therefore, the dry state evaluation index Cdry is calculated from the operation parameter having various units using the relationship of FIG. The dry state evaluation index Cdry is for converting an operation parameter having various units into a common index representing the degree of dry state of the electrolyte membrane. Since the range and numerical value of the dry state evaluation index Cdry can be arbitrarily set, a positive value range that is easy to handle and a specific numerical value [anonymous number] may be determined. Here, it is assumed that the degree of dryness of the electrolyte membrane is larger as the numerical value of the dryness evaluation index Cdry is larger in positive value.

  The ratio T3 / (T1 + T2 + T3) of pressure increase per unit cycle is set from the dry state evaluation index Cdry numerically expressed as a positive value in this way using the relationship shown in FIG. Basically, the higher the value of the dry state evaluation index Cdry is, the larger the value is, and the higher the rate of pressure increase per unit cycle.

  However, it is characteristic that the drying of the anode gas flow path 46 on the cell stack 2 outlet side (the cathode gas flow path 45 on the cell stack 2 inlet side) proceeds rapidly from a certain point as the operating parameters change. In consideration of this point, the dry state evaluation index Cdry and the rate of pressure increase per unit cycle do not simply have a proportional relationship. That is, the rate of pressure increase per unit cycle is constant until the dry state evaluation index Cdry reaches the first threshold C1, and when the dry state evaluation index Cdry reaches the first threshold C1, The step-up rate is increased stepwise (see the solid line in FIG. 8). Also, a second threshold value C2 (C1 <C2) larger than the first threshold value C1 is provided, and when the second threshold value C2 is reached, the ratio of boosting per unit cycle is further increased stepwise (FIG. (See solid line 8). Of course, detection errors (estimation errors) and detection time delays (estimation time delays) may occur when detecting (or estimating) the operating parameters shown in <1> to <10> above. In order to avoid the deterioration of the determination accuracy, the rate of pressure increase per unit cycle is made smoother from the point at which the dry state evaluation index Cdry has reached the third threshold C3 (C3 <C1) as shown by the one-dot chain line in FIG. It may be increased.

  In other words, FIG. 8 means that it is handled as follows. That is, according to the characteristic of the solid line in FIG. 8, when the dry state evaluation index Cdry is less than the first threshold C1, the electrolyte membrane is in a wet state, and the dry state evaluation index Cdry is equal to or greater than the first threshold C1. The electrolyte membrane is considered to be in a dry state. Furthermore, the dry state is divided into two when the first threshold C1 or more and less than the second threshold, and when the dry state is the second threshold C2 or more, and the dry state evaluation index Cdry is the first threshold C1 or more. If the electrolyte membrane is in the dry state 1 (ie, the degree of drying is relatively small) when it is less than the threshold value of 2, the electrolyte membrane is dried when the dry state evaluation index Cdry is equal to or greater than the second threshold value C2. It is considered to be in state 2 (a state where the degree of drying is relatively large). Here, the electrolyte membrane considering whether it is in a dry state or a wet state is particularly an electrolyte membrane on the cell stack 2 outlet side of the anode gas flow channel 46 (cell stack 2 inlet side of the cathode gas flow channel 45). That's it.

  On the other hand, according to the characteristics of the one-dot chain line in FIG. 8, when the dry state evaluation index Cdry is less than the third threshold C3, the electrolyte membrane is in a wet state, and the dry state evaluation index Cdry is the third threshold C3. When it is above, it is considered that the electrolyte membrane is in a dry state.

  As described above, FIG. 8 shows two examples of how the dry state is expressed: a case where it is expressed as a discrete value (solid line) and a case where it is expressed as a continuous value (dashed line). It is not limited to. For example, instead of dividing the degree of drying into two, the degree of drying can be further divided into three or more.

  In the above formula (2), in order to increase the rate of boosting per unit period, the boost time T3 that is the numerator on the right side of the formula (2) is increased, or the time that is the denominator on the right side of the formula (2) is shortened. That's fine. In order to shorten the time that is the denominator of the right side of the equation (2), the decompression time T2 may be shortened or the maintenance time T1 may be shortened. Here, if the concept of “increasing the rate of boosting per unit cycle” as in the above-described embodiment is a superordinate concept, “increasing the boosting time T3” and “decreasing the sustaining time T1”. “To shorten the decompression time T2” corresponds to a subordinate concept. Therefore, next, this subordinate concept will be described in the first to fifth embodiments. That is, the case of “increasing the pressure increase time T3” is the first, second and third embodiments, the case of “decreasing the maintenance time T1” is the case of the fourth embodiment, and the case of “decreasing the pressure reduction time T2”. Will be described in a fifth embodiment.

  FIG. 9 is a timing chart showing changes in the operating state of the hydrogen pressure regulating valve 5, the anode gas flow path pressure, and the anode gas flow in a unit cycle in the anode dead end operation of the first embodiment. In addition, the time when it is determined that the electrolyte membrane is in a wet state is indicated by a one-dot chain line, and the time when it is determined that the electrolyte membrane is in a dry state is indicated by a solid line. For comparison, FIG. 6 shows changes in the operating state, the anode gas flow path pressure, and the anode gas flow of the hydrogen pressure regulating valve 5 in a unit cycle in the anode dead end operation of the prior art.

  In the first embodiment, as shown in the middle part of FIG. 9, when it is determined that the electrolyte membrane is in a dry state, the pressure increase time T3 is set longer than when it is determined that the electrolyte membrane is in a wet state. As a result, as shown in the lower part of FIG. 9, when the electrolyte membrane is determined to be in a dry state, the time during which the supplied fuel gas flows in a forward flow is longer than when it is determined that the electrolyte membrane is in a wet state. As a result, the amount of water vapor that moves from the upstream side to the downstream side of the anode gas channel 46 is greater than when the electrolyte membrane is in a wet state. Thus, even when it is determined that the electrolyte membrane is in a dry state, humidification of the electrolyte membrane on the cell stack 2 outlet side of the anode gas channel 46 (on the cell stack 2 inlet side of the cathode gas channel 45) can be realized. .

  In FIG. 9, the unit period in the anode dead end operation is the time (maintenance time) T1 during which the pressure in the anode gas flow path 46 inside the cell stack 2 is maintained at the upper limit pressure P1, and the anode gas flow path inside the cell stack 2. The time (pressure reduction time) T2 required for the process of reducing the pressure of 46 from the upper limit pressure P1 to the lower limit pressure P2, and the pressure in the anode gas flow path 46 inside the cell stack 2 is increased from the lower limit pressure P2 to the upper limit pressure P1. This is the total of three times (= T1 + T2 + T3) with the time (step-up time) T3 required for the process. By increasing the boosting time T3, the ratio of boosting per unit cycle (T3 / (T1 + T2 + T3)) is increased.

  As a measure for extending the pressure increase time T3 when it is determined that the electrolyte membrane is in a wet state when it is determined that the electrolyte membrane is in a dry state, a hydrogen pressure regulating valve 5 (fuel gas flow rate adjusting means) is used. It is desirable to reduce the flow rate of the supplied fuel gas (the supplied fuel gas flow rate per unit time) by adjusting the opening. That is, as shown in the upper part of FIG. 9, the hydrogen pressure regulating valve when the electrolyte membrane is determined to be in a dry state rather than the opening degree a of the hydrogen pressure regulating valve 5 when it is determined that the electrolyte membrane is wet. 5 is reduced.

  As described above, when the flow rate of the supplied fuel gas is reduced, the high-humidity exhaust cathode gas (the high-humidity gas exhausted from the cathode at the outlet side of the cell stack 2) enters the cell stack 2 inlet of the anode gas flow path 46. The efficiency of exchanging humidity with the supplied fuel gas on the side increases compared with the case where the flow rate of the supplied fuel gas is not reduced. Therefore, the relative humidity of the supplied fuel gas at the cell stack 2 outlet side of the anode gas flow path 46 increases, and at the cell stack 2 outlet side of the anode gas flow path 46 (cell stack 2 inlet side of the cathode gas flow path 45). The effect of humidifying the electrolyte membrane can be enhanced.

In order to sufficiently increase the humidity of the anode gas on the cell stack 2 outlet side of the anode gas flow path 46, the pressure increase time T3 that affects the flow rate of the supplied fuel gas is:
T3> (Ps / P1) × T2 (3)
Where Ps; saturation vapor pressure of anode gas (calculated from cell stack temperature),
P1: upper limit pressure of anode gas flow path in anode dead end operation,
T2: decompression time,
It is desirable to set so as to satisfy the relationship.

  Explaining this, the flow rate of the supplied fuel gas can be lowered to a level at which the anode gas at the outlet side of the cell stack 2 in the anode gas flow path 46 can be brought close to the humidity of 100%. As the saturated vapor pressure in the anode gas increases, the relative humidity is less likely to increase. Therefore, it is desirable to increase the pressure increase time T3 and decrease the flow rate of the supplied fuel gas per unit time. Further, as the pressure of the supplied fuel gas is lower, the volume flow rate of the supplied fuel gas increases and the relative humidity is less likely to increase. Therefore, it is desirable to lengthen the pressure increase time T3 and lower the supplied fuel gas flow rate per unit time.

  Further, the supply fuel gas flow rate at the time of pressure increase in the anode dead end operation is proportional to the consumption amount of the anode gas at the time of pressure reduction in the immediately preceding anode dead end operation. Therefore, it is effective to determine the pressure increase time T3 in proportion to the length of the pressure reduction time T2.

Putting these together,
T3∝ {(saturated vapor pressure of anode gas) / (supply fuel gas pressure)} × T2
... (Supplement 1)
The relationship is obtained. (Supplement 1) The supply fuel gas pressure of the formula (1) is the upper limit of the period (unit period) of pressure pulsation comprising the process of increasing the pressure of the anode gas flow path 46 in the cell stack 2 and the process of decreasing the pressure It is most effective to use the pressure P1 as a representative value.

As a result, the formula (A1) is rewritten and T3∝ (Ps / P1) × T2 (A2)
Where Ps; saturation vapor pressure of anode gas (calculated from cell stack temperature),
P1: upper limit pressure of anode gas flow path in anode dead end operation,
T2: decompression time,
The following equation is obtained. The above equation (3) was obtained from this (complement 2) equation.

  In this way, the pressure increase time T3 is determined from the above equation (3), and the steam exchange efficiency of the anode gas on the outlet side of the cell stack 2 in the anode gas flow path 46 is increased, so that the cell stack 2 in the anode gas flow path 46 is increased. High humidity gas can be supplied from the outlet side.

  Specific numerical values for the above equation (3) are given when the upper limit pressure P1 of the anode gas flow path 46 is 120 kPa (absolute pressure) and the saturated vapor pressure of the anode gas calculated from the temperature of the cell stack 2 is 30 kPa. Is Ps / P1 = 30/120 = 25%. At this time, it is effective that the pressure increase time T3 exceeds 25% of the pressure decrease time T2 from the above equation (3). For example, when the pressure reduction time T2 is 10 seconds, it is desirable that the pressure increase time T3 exceeds 2.5 seconds (= 10 seconds × 25%).

  In order to effectively incorporate water vapor retained in the anode gas into the membrane of the electrolyte membrane, it is effective to determine the pressure increase time T3 in consideration of the diffusion time of water in the membrane. That is, it is desirable that the pressure increase time T3 is longer than the water diffusion time Td in the electrolyte membrane.

Here, the water diffusion time Td is obtained by using the thickness L of the electrolyte membrane and the diffusion coefficient D of water in the electrolyte membrane,
Td = L ² / D (4)
It is calculated by the following formula.

The water vapor contained in the anode gas on the outlet side of the cell stack 2 in the anode gas flow path 46 moves from the anode side to the cathode side by diffusion through the electrolyte membrane. Therefore, the time required for the electrolyte membrane to absorb water vapor from the anode gas on the outlet side of the cell stack 2 of the anode gas flow path 46 can be regarded as the same as the water diffusion time Td in the above equation (4). Therefore, when it is determined that the electrolyte membrane is in a dry state in this way, the pressure increase time T3 is set to the water diffusion time L ² / calculated from the film thickness L of the electrolyte membrane and the diffusion coefficient D of water in the electrolyte membrane. By making it longer than D, water vapor held in the anode gas on the outlet side of the cell stack 2 of the anode gas flow path 46 can be effectively taken into the membrane of the electrolyte membrane.

  When specific numerical values are given for the above formula (4), in the case of a general Nafion 211 (film thickness L is 25 μm) as an electrolyte membrane of a polymer electrolyte fuel cell, the water diffusion time Td in the electrolyte membrane is about 2 seconds. It is. Therefore, when using Nafion 211 (film thickness L is 25 μm), it is desirable that the pressure increase time T3 exceeds 2 seconds. In this case, in order to satisfy the above expression (3), it is desirable that the boosting time T3 exceeds 2.5 seconds. “Nafion” is a registered trademark of DuPont.

  On the other hand, since the boosting time T3 is preferably shorter than the decompression time T2 (for example, about 10 seconds), the boosting time T3 is desirably in the range of 2.5 to 10 seconds.

  When it is determined that the electrolyte membrane is in a dry state, the pressurization time T3 is switched to a range of 2.5 to 10 seconds, and it is determined that the electrolyte membrane is again in a dry state even after the pressurization time T3 is thus increased. In this case (that is, when drying of the electrolyte membrane further proceeds), it is most effective to increase the pressure increase time T3 within the above range.

  Here, the depressurization time T2 may be replaced with a time for fully closing the hydrogen pressure regulating valve 5 as shown in the upper and middle stages of FIG.

When performing control to slightly open the hydrogen pressure control valve 5 during pressure reduction during anode dead end operation and slow down pressure reduction,
T2eff = T2 × (1−hydrogen pressure regulating valve opening) (5)
However, T2eff; substantial decompression time,
The substantial decompression time T2eff may be calculated by the following formula, and the substantial decompression time T2eff may be regarded as the decompression time T2 anew.

  Here, when the hydrogen pressure regulating valve 5 is fully closed, since the hydrogen pressure regulating valve opening degree = 0, the substantial pressure reducing time T2eff is equal to the pressure reducing time T2 from the equation (5). On the other hand, if the hydrogen pressure regulating valve 5 is slightly opened, the hydrogen pressure regulating valve opening becomes a small value that is not zero, so that the substantial pressure reducing time T2eff is shorter than the pressure reducing time T2.

  FIG. 10 is a timing chart showing changes in the operating state of the hydrogen pressure regulating valve 5, the anode gas flow path pressure, and the anode gas flow in a unit cycle in the anode dead end operation of the second embodiment. Also in the second embodiment, the time when it is determined that the electrolyte membrane is in a wet state is indicated by an alternate long and short dash line, and the time when it is determined that the electrolyte membrane is in a dry state is indicated by a solid line.

  In the second embodiment, as shown in the middle part of FIG. 10, when it is determined that the electrolyte membrane is in a dry state, the upper limit pressure P1 in the anode dead end operation using the hydrogen pressure regulating valve 5 (upper limit pressure adjusting means). 'Is made higher than the upper limit pressure P1 in the anode dead-end operation when it is determined that the electrolyte membrane is in a wet state.

  In the second embodiment, the depressurization time T2 and the pressurization time T3 increase in proportion to the amount of increase in the upper limit pressure (= P1′−P1) from when it is determined that the electrolyte membrane is in a wet state. When it is determined that the membrane is in a dry state, the rate of pressure increase per unit cycle is greater than when it is determined that the electrolyte membrane is in a wet state. That is, according to the second embodiment, as shown in the lower part of FIG. 10, when it is determined that the electrolyte membrane is in a dry state, the supplied fuel gas is more forward than when it is determined that the electrolyte membrane is in a wet state. The flowing time becomes longer. As a result, the amount of water vapor transferred from the upstream side to the downstream side of the anode gas channel 46 increases. Therefore, it becomes possible to humidify the electrolyte membrane on the cell stack 2 outlet side of the anode gas channel 46 (on the cell stack 2 inlet side of the cathode gas channel 45), and on the cell stack 2 inlet side of the cathode gas channel 45 It is possible to prevent a decrease in power generation performance due to drying of the electrolyte membrane.

  However, compared with the first embodiment, the second embodiment does not provide an effect of lowering the flow rate of the supplied fuel gas at the time of pressure increase in the anode dead end operation. The effect of increasing the humidity exchange efficiency to the anode gas cannot be obtained.

  FIG. 11 is a schematic configuration diagram of the fuel cell system according to the third embodiment. In addition to the first embodiment, a water amount sensor 54 for detecting the amount of liquid water stored in the water separator tank 7 is additionally provided. . The same parts as those in FIG. 1 of the first embodiment are denoted by the same reference numerals.

  The third embodiment determines that the electrolyte membrane is in a dry state when the liquid water amount Qw in the water separator tank 7 (gas storage means) detected by the water amount sensor 54 is less than the upper limit value Q2 of the liquid water amount. Sometimes, the drain valve 9 (volume adjustment mechanism) is opened to discharge liquid water to the outside, and the volume of the gas phase portion in the water separator tank 7 is made larger than when it is determined that the electrolyte membrane is in a wet state. Thus, the pressure increase time T3 is lengthened, thereby improving the drainage performance from the anode gas flow path 46 inside the cell stack 2 to the discharge pipe 6 and the water separator tank 7.

  This principle will be described. In the state where the upper limit pressure P1 is set by the hydrogen pressure regulating valve 5, the supplied fuel gas is allowed to flow in the forward direction to raise the anode gas flow path 46 in the cell stack 2 to the upper limit pressure P1. Considering two cases where the volume of the anode gas flow path flowing to the drain valve 9 is relatively large and relatively small, both supply fuel gas flowing from the hydrogen pressure regulating valve 5 in the forward direction is considered. Suppose the mass flow is equal. At this time, the volume of the anode gas flow path flowing from the hydrogen pressure regulating valve 5 to the drain valve 9 is larger when the volume of the anode gas flow path flowing from the hydrogen pressure regulating valve 5 to the drain valve 9 is relatively large. The time (that is, the pressure increase time T3) is longer until the anode gas flow path 46 in the cell stack 2 reaches the upper limit pressure P1 than when the pressure is relatively small. For this reason, when the volume of the anode gas flow path flowing from the hydrogen pressure regulating valve 5 to the drain valve 9 is relatively large, it takes a longer time than the liquid water on the anode gas flow path 46 inside the cell stack 2. The dynamic pressure toward the downstream side can be applied. As a result, when the volume of the anode gas flow path flowing from the hydrogen pressure regulating valve 5 to the drain valve 9 is relatively large, the discharge pipe 6 and the water separator tank are connected from the anode gas flow path 46 inside the cell stack 2. The drainage to 7 will be improved.

  As described above, according to the third embodiment as well, when the electrolyte membrane is determined to be in a dry state, the time during which the supplied fuel gas flows in a forward flow is longer than when it is determined that the electrolyte membrane is in a wet state. As a result, the amount of water vapor transferred from the upstream side to the downstream side of the anode gas channel 46 increases. Therefore, it becomes possible to humidify the electrolyte membrane on the cell stack 2 outlet side of the anode gas channel 46 (on the cell stack 2 inlet side of the cathode gas channel 45), and on the cell stack 2 inlet side of the cathode gas channel 45 It is possible to prevent a decrease in power generation performance due to drying of the electrolyte membrane.

  The third embodiment is not limited to this, and may have a structure provided with a volume adjustment mechanism that can adjust the volume of the water separator tank 7 regardless of the amount of liquid water stored in the water separator tank 7. . In this case, when the liquid water amount Qw in the water separator tank 7 detected by the water amount sensor 54 is less than the upper limit value Q2 of the liquid water amount, when it is determined that the electrolyte membrane is in a dry state, the volume adjustment mechanism is By using the volume of the water separator tank 7 to be larger than when it is determined that the electrolyte membrane is in a wet state, the pressure increase time T3 is lengthened. Thereby, the drainage property from the anode gas flow path 46 inside the cell stack 2 to the discharge pipe 6 and the water separator tank 7 can be improved.

  Instead of the water amount sensor, a water level sensor that detects the level of liquid water in the water separator tank 7 can be used.

  Control performed by the controller 51 of the third embodiment will be described with reference to FIG. FIG. 12 is for opening and closing the drain valve 9 and is executed at regular intervals (for example, every 10 ms). The flow shown in FIG. 12 is executed with the start of the anode dead end operation or immediately before the start of the anode dead end operation.

  In step 1, the liquid water amount Qw in the water separator tank 7 detected by the water amount sensor 54 and the dry state evaluation index Cdry are read. Here, the dry state evaluation index Cdry is calculated by referring to the above FIG. 7 from any one of the operation parameters <1> to <10> described above.

  Step 2 looks at the dry state flag. This flag is initially set to zero. Now, proceed to Steps 3 and 4 assuming that the dry state flag = 0. In step 3, the liquid water amount Qw in the water separator tank 7 and the upper limit value Q2 of the liquid water amount in the water separator tank 7 are set, and in step 4, the dry state evaluation index Cdry and a threshold value for determining that the liquid is in the dry state (FIG. 7). If the characteristic of the solid line shown in FIG. 2 is used, it is compared with the first threshold value C1). Here, the upper limit Q2 of the amount of liquid water in the water separator tank 7 is determined in advance from the specifications of the water separator tank 7.

  When the liquid water amount Qw in the water separator tank 7 is less than the upper limit value Q2 of the liquid water amount in the water separator tank 7 and the dry state evaluation index Cdry is not less than the threshold value C1, it is determined that the electrolyte membrane is in a dry state, Proceeding to step 5, the dry state flag = 1 is set. In step 6, in order to increase the volume of the gas phase portion in the water separator tank 7, the drain valve 9 is opened. Thereby, the liquid water in the water separator tank 7 is discharged to the outside, and an operation for increasing the volume of the gas phase portion in the water separator tank 7 is started.

  On the other hand, when the dry state evaluation index Cdry is less than the first threshold value C1 in step 4, it is determined that the electrolyte membrane is in a wet state, and the process proceeds to step 7 to keep the drain valve 9 in a fully closed state.

  When the amount Qw of liquid water in the water separator tank 7 is equal to or greater than the upper limit value Q2 of the amount of liquid water in the water separator tank 7 in step 3, the flow proceeds to step 8 and the drain valve 9 is immediately opened to liquid water in the water separator tank 7 To the outside. This is due to the following reason. That is, it is necessary to reversely flow the anode gas stored in the water separator tank toward the anode gas flow path 46 in the cell stack 2 at the time of pressure reduction in the anode dead end operation. However, if liquid water of the upper limit value Q2 or more is stored in the water separator tank 7, it is conceivable that the anode gas that flows backward toward the anode gas flow path 46 inside the cell stack 2 is insufficient. Therefore, in order to prevent a shortage of anode gas that flows back toward the anode gas flow path 46 in the cell stack 2 during decompression in the anode dead end operation, liquid water having an upper limit value Q2 or more is stored in the water separator tank 7. This is because the liquid water in the water separator tank 7 is discharged to ensure a certain amount of anode gas in the water separator tank 7.

  When the dry state flag is set to 1 in step 5 last time, the process proceeds from step 1 to step 9 to step 9 from the next time, and the dry state evaluation index Cdry is compared with the first threshold value C1. Immediately after the dry state flag = 1, the dry state evaluation index Cdry is equal to or greater than the first threshold value C1, and therefore, the process proceeds to step 10 and the liquid water amount Qw in the water separator tank 7 and the liquid water amount in the water separator tank 7 are set. The lower limit value Q1 is compared. Here, a lower limit value Q1 of the amount of liquid water in the water separator tank 7 is also determined in advance. When the liquid water amount Qw in the water separator tank 7 is not less than the lower limit value Q1 of the liquid water amount in the water separator tank 7, the process proceeds to step 6 and the operation of step 6 is executed. That is, the drain valve 9 is continuously opened, and the liquid water in the water separator tank 7 is discharged.

  When the liquid water amount Qw in the water separator tank 7 eventually becomes less than the lower limit value Q1 of the liquid water amount in the water separator tank by repeating Steps 1, 2, 9, 10, and 6, the process proceeds to Step 11 and all the drain valves 9 are turned on. Close. Even in this state, as long as the dry state evaluation index Cdry is not less than the first threshold value C1 in step 9, the operations of steps 1, 2, 9, 10, and 11 are repeated.

  In this way, when the electrolyte membrane is in a dry state during the anode dead end operation, liquid water is discharged from the drain valve 9 and the volume of the gas phase portion in the water separator tank 7 is wet. (See steps 2, 5, 6, step 2, 9, 10, 6, step 2, 9, 10, 11 in FIG. 12).

  By increasing the volume of the gas phase portion of the water separator tank 7, the pressure increase time T <b> 3 is lengthened, whereby the drainage performance from the anode gas flow path 46 inside the cell stack 2 to the discharge pipe 6 and the water separator tank 7 is improved. improves. As a result, in step 9, the dry state evaluation index Cdry is less than the first threshold value C1, and it is determined that the electrolyte membrane has become wet. At this time, the process proceeds from step 9 to steps 12 and 13 where the dry state flag = 0 and the drain valve 9 is fully closed.

  FIG. 13 is a timing chart showing changes in the operating state of the hydrogen pressure regulating valve 5, the anode gas flow path pressure, and the anode gas flow in a unit cycle in the anode dead end operation of the fourth embodiment. Also in the fourth embodiment, the time when it is determined that the electrolyte membrane is in a wet state is indicated by an alternate long and short dash line, and the time when it is determined that the electrolyte membrane is in a dry state is indicated by a solid line.

  During the maintenance time T1 in the anode dead end operation, the flow rate of the supplied fuel gas from the upstream side to the downstream side of the anode gas flow path 46 is zero, and the water vapor moves from the upstream side to the downstream side of the anode gas flow path 46. Therefore, drying of the electrolyte membrane easily proceeds on the cell stack 2 outlet side of the anode gas flow path 46 (on the cell stack 2 inlet side of the cathode gas flow path 45). That is, while maintaining a constant pressure, only the supply fuel gas corresponding to the anode gas consumption accompanying power generation is supplied to the anode, and the flow rate of the supply fuel gas decreases toward the downstream side of the anode gas flow path 46. For this reason, the amount of water vapor that can be transported from the upstream side to the downstream side of the anode gas channel 46 decreases as it goes to the downstream side of the anode gas channel 46. During a low load operation, the amount of water vapor taken away by the cathode gas is greater on the downstream side of the anode gas flow channel 46 than the amount of water vapor transported from the upstream side of the anode gas flow channel 46, so Drying of the electrolyte membrane proceeds on the side (the cell stack 2 inlet side of the cathode gas flow channel 45).

  Therefore, in the fourth embodiment, as shown in the middle part of FIG. 13, when it is determined that the electrolyte membrane is in a dry state, the maintenance time T1 in the anode dead end operation is determined to be that the electrolyte membrane is in a wet state. The maintenance time T1 is shortened from time to time.

  Thus, if the maintenance time T1 is shortened when it is determined that the electrolyte membrane is in a dry state, the pressure increase time T3 is not different from that when it is determined that the electrolyte membrane is in a wet state, but the pressure per unit cycle Since the rate of maintenance decreases, the rate of pressure increase per unit cycle is greater than when it is determined that the electrolyte membrane is in a wet state. That is, in the fourth embodiment, as shown in the lower part of FIG. 13, when it is determined that the electrolyte membrane is in a dry state, the supplied fuel per unit cycle is greater than when it is determined that the electrolyte membrane is in a wet state. The proportion of time during which the gas flows in the forward flow increases. As a result, the amount of water vapor transferred from the upstream side to the downstream side of the anode gas flow path 46 increases as the proportion of the time during which the supplied fuel gas flows in a forward flow per unit period increases. Therefore, it becomes possible to humidify the electrolyte membrane on the cell stack 2 outlet side of the anode gas channel 46 (on the cell stack 2 inlet side of the cathode gas channel 45), and on the cell stack 2 inlet side of the cathode gas channel 45 It is possible to prevent a decrease in power generation performance due to drying of the electrolyte membrane.

  FIG. 14 is a timing chart showing changes in the operating state of the hydrogen pressure regulating valve, the anode gas flow path pressure, and the anode gas flow in a unit cycle in the anode dead end operation of the fifth embodiment. Also in the fifth embodiment, the time when it is determined that the electrolyte membrane is in a wet state is indicated by an alternate long and short dash line, and the time when it is determined that the electrolyte membrane is in a dry state is indicated by a solid line.

  At the time of depressurization in the anode dead end operation, the backflow of the anode gas from the water separator tank 7 toward the anode gas flow path 46 inside the cell stack 2 occurs due to consumption of the anode gas accompanying power generation. During low load operation, only a small amount of water vapor in the discharge pipe 6 and the water separator tank 7 returns to the outlet side of the cell stack 2 in the anode gas passage 46, and the outlet side of the cell stack 2 in the anode gas passage 46 (cathode gas). The cell stack 2 inlet side of the flow path 45 is easily dried.

  Therefore, in the fifth embodiment, as shown in the middle part of FIG. 14, when it is determined that the electrolyte membrane is in the dry state, the pressure reduction time T2 in the anode dead end operation is determined as the electrolyte membrane is in the wet state. Make it shorter.

  As described above, when the depressurization time T2 is shortened when it is determined that the electrolyte membrane is in a dry state, the pressurization time T3 is not different from that when it is determined that the electrolyte membrane is in a wet state. Since the rate of depressurization decreases, the rate of pressure increase per unit cycle is greater than when it is determined that the electrolyte membrane is in a wet state. That is, in the fifth embodiment, as shown in the lower part of FIG. 14, when it is determined that the electrolyte membrane is in a dry state, the supplied fuel per unit cycle is greater than when it is determined that the electrolyte membrane is in a wet state. The proportion of time during which the gas flows in the forward flow increases. As a result, the amount of water vapor transferred from the upstream side to the downstream side of the anode gas flow path 46 increases as the proportion of the time during which the supplied fuel gas flows in a forward flow per unit period increases. Therefore, it is possible to humidify the electrolyte membrane on the cell stack side of the anode gas channel 46 (on the cell stack 2 inlet side of the cathode gas channel 45), and the electrolyte on the cell stack 2 inlet side of the cathode gas channel 45 A decrease in power generation performance due to the drying of the membrane can be prevented.

  The pressure reduction time T2 becomes shorter as the amount of anode gas consumed by power generation increases and the amount of gas discharged from the nitrogen purge valve 11 increases. Therefore, when it is determined that the electrolyte membrane is in a dry state during pressure reduction in anode dead-end operation In addition, the opening degree of the nitrogen purge valve 11 (exhaust valve) at the time of depressurization in the anode dead end operation may be made larger than when it is determined that the electrolyte membrane is in a wet state.

  As described above, when it is determined that the electrolyte membrane is in a dry state, the opening degree of the nitrogen purge valve 11 (exhaust valve) at the time of depressurization in the anode dead end operation is made larger than when it is determined that the electrolyte membrane is in a wet state. The anode gas channel 46 is quickly depressurized and the depressurization time T2 is shortened. As a result, when it is determined that the electrolyte membrane is in a dry state, the ratio of the time during which the supplied fuel gas flows in a forward flow per unit cycle is larger than when it is determined that the electrolyte membrane is in a wet state. The amount of water vapor transferred from the upstream side to the downstream side of the minute anode gas channel 46 increases. Therefore, it is possible to humidify the electrolyte membrane at the cell stack outlet of the anode gas channel 46 (the cell stack 2 inlet side of the cathode gas channel 45), and the electrolyte at the cell stack 2 inlet side of the cathode gas channel 45 A decrease in power generation performance due to the drying of the membrane can be prevented.

  When it is determined that the electrolyte membrane is in a dry state, the ratio of pressure increase per unit cycle may be increased by combining two or more of the first to fifth embodiments. For example, when the pressure reduction time T2 is shortened by increasing the opening of the nitrogen purge valve 11 as in the fifth embodiment, the anode gas in the gas discharged to the discharge pipe 6 increases and fuel is used. The rate gets worse. In this case, as shown by the solid line in FIG. 8, the dry state is divided into two, and in the dry state 1 where the dry state evaluation index Cdry is less than the second threshold C2, the first to fourth Any one of the embodiments may be employed, and the fifth embodiment may be employed only in the dry state 2 in which the dry state evaluation index Cdry is equal to or greater than the second threshold C2.

  In FIG. 9, FIG. 10, FIG. 13, and FIG. 14, for simplicity, there are two cases: when it is determined that the electrolyte membrane is in a dry state, and when it is determined that the electrolyte membrane is in a wet state. (Ie binary). 9, 10, 13, and 14, when it is determined that the electrolyte membrane is in a wet state, the dry state evaluation index Cdry is less than the first threshold value C <b> 1 according to the characteristics of the solid line in FIG. 8. When the dry state evaluation index Cdry is less than the third threshold C3 according to the characteristics of the one-dot chain line in FIG. Similarly, when it is determined that the electrolyte membrane is in a dry state, according to the characteristic of the solid line in FIG. 8, when the dry state evaluation index Cdry is equal to or higher than the first threshold value C1, the characteristic of the one-dot chain line in FIG. According to this, the dry state evaluation index Cdry is equal to or greater than the third threshold value C3.

  However, the present invention is not limited to the cases shown in FIGS. 9, 10, 13, and 14. For example, the operation state of the hydrogen pressure regulating valve 5 in the unit cycle in the anode dead-end operation when the solid line characteristic of FIG. 8 is adopted and the dry state is divided into the dry state 1 and the dry state 2, the anode gas A timing chart showing changes in the flow path pressure and the anode gas flow is as shown in FIG. 15 instead of FIG. 9 of the first embodiment. In FIG. 15 showing another example of the first embodiment, the time when the electrolyte membrane is determined to be in the dry state 1 is indicated by a broken line, the time when the electrolyte membrane is determined to be in the dry state 2 is indicated by a solid line, and the electrolyte membrane is wet. The time when it is determined to be in a state is indicated by a one-dot chain line.

  In another example of the first embodiment, as shown in the middle part of FIG. 15, when it is determined that the electrolyte membrane is in the dry state 1, the pressure increase time T <b> 3 is greater than when the electrolyte membrane is determined to be in the wet state. Lengthen. As shown in the middle part of FIG. 15, when it is determined that the denatured film is in the dry state 2, the pressurization time T <b> 3 is set longer than when it is determined that the electrolyte film is in the dry state 1.

  As a result, as shown in the lower part of FIG. 15, when the electrolyte membrane is determined to be in the dry state 1, the time for the supplied fuel gas to flow in the forward flow becomes longer than when the electrolyte membrane is determined to be in the wet state. When it is determined that the electrolyte membrane is in the dry state 2, the time during which the supplied fuel gas flows in the forward flow is further longer than when it is determined that the electrolyte membrane is in the dry state 1. As a result, the amount of water vapor that moves from the upstream side to the downstream side of the anode gas channel 46 is greater than when it is determined that the electrolyte membrane is in a wet state. Thus, humidification of the electrolyte membrane on the cell stack 2 outlet side of the anode gas flow path 46 (cell stack 2 inlet side of the cathode gas flow path 45) can be realized according to the degree of drying.

DESCRIPTION OF SYMBOLS 1 Fuel cell system 2 Cell stack 5 Hydrogen pressure regulation valve (fuel gas flow rate adjustment means, upper limit pressure adjustment means)
7 Water separator tank (gas storage means)
9 Drain valve (volume adjustment mechanism)
11 Nitrogen purge valve (exhaust valve)
46 Anode gas flow path 51 Controller

Claims (12)

A cell stack in which a plurality of unit fuel cells configured by sandwiching an electrolyte membrane between an anode and a cathode is provided,
An operation of supplying fuel gas to the anode so that a unit period is repeated, with a period of pressure pulsation consisting of a process of increasing and decreasing a pressure of the anode gas flow path inside the cell stack as a unit period. In the fuel cell system to perform,
Determining means for determining whether the electrolyte membrane is in a dry state or a wet state during the operation;
A fuel cell system comprising: a pressure increase ratio changing means for increasing the rate of pressure increase per unit cycle when the electrolyte membrane is in a dry state based on the determination result, as compared with when the electrolyte membrane is in a wet state. The unit period is the sum of a pressure increase time that is a time required for the pressure increasing process and a pressure reduction time that is a time required for the pressure decreasing process,
2. The fuel cell system according to claim 1, wherein when the electrolyte membrane is in a dry state, the pressurization time is set longer than when the electrolyte membrane is in a wet state. A fuel gas flow rate adjusting means capable of adjusting the flow rate of the fuel gas supplied to the anode;
When the electrolyte membrane is in a dry state, the fuel gas flow rate adjusting means is used to make the flow rate of the fuel gas supplied to the anode smaller than when the electrolyte membrane is in a wet state, thereby increasing the pressure increase time. The fuel cell system according to claim 2, wherein: An upper limit pressure adjusting means capable of adjusting an upper limit pressure at which the pressure of the anode gas channel reaches in the process of increasing the pressure;
When the electrolyte membrane is in a dry state, the pressurization time is lengthened by making the upper limit pressure larger than when the electrolyte membrane is in a wet state using the upper limit pressure adjusting means. Item 3. The fuel cell system according to Item 2. A gas storage means for storing a gas discharged to the outside of the cell stack without being consumed at the anode;
A volume adjustment mechanism capable of adjusting the volume of the gas storage means,
When the electrolyte membrane is in a dry state, the volume adjustment mechanism is used to make the volume of the gas storage means larger than when the electrolyte membrane is in a wet state, thereby extending the pressurization time. The fuel cell system according to claim 2. A gas storage means for storing a gas discharged to the outside of the cell stack without being consumed at the anode;
A drain valve for opening and closing a passage communicating with the atmosphere at the lower part of the gas storage means,
3. The fuel cell according to claim 2, wherein when the electrolyte membrane is in a dry state, the pressure increase time is lengthened by opening the drain valve and discharging the liquid water in the gas storage means. system.   When the electrolyte membrane is in a dry state, the pressurization time is longer than the water diffusion time calculated from the thickness of the electrolyte membrane and the diffusion coefficient of water in the electrolyte membrane. Item 7. The fuel cell system according to any one of Items 2 to 6. When the electrolyte membrane is in a dry state, the pressure increase time and the pressure reduction time are:
T3> (Ps / P1) × T2
However, T3; said pressure | voltage rise time,
T2: the decompression time,
Ps; saturation vapor pressure of anode gas,
P1; upper limit pressure of the pressure pulsation,
The fuel cell system according to any one of claims 2 to 7, wherein the relationship is satisfied. Adding a process in which the pressure of the anode gas flow path inside the cell stack is kept constant between the process of increasing pressure and the process of reducing pressure;
The unit period is a sum of a pressure increase time that is a time required for the pressure increasing process, a pressure reduction time that is a time required for the pressure decreasing process, and a maintenance time that is a time required for the additional maintained process. And
The rate of pressure increase per unit cycle is increased by shortening the maintenance time when the electrolyte membrane is in a dry state than when the electrolyte membrane is in a wet state. The fuel cell system described in 1.   When the electrolyte membrane is in a dry state, the rate of pressure increase per unit cycle is increased by shortening the decompression time, which is the time required for the decompression process, compared to when the electrolyte membrane is in a wet state. The fuel cell system according to claim 1, wherein: A gas storage means for storing a gas discharged to the outside of the cell stack without being consumed at the anode;
An exhaust valve for opening and closing a passage communicating with the atmosphere at the top of the gas storage means,
The rate of pressure increase per unit cycle is increased by increasing the opening of the exhaust valve when the electrolyte membrane is in a dry state than when the electrolyte membrane is in a wet state. Item 4. The fuel cell system according to Item 1. A cell stack in which a plurality of unit fuel cells configured by sandwiching an electrolyte membrane between an anode and a cathode is provided,
An operation of supplying fuel gas to the anode so that a unit period is repeated, with a period of pressure pulsation consisting of a process of increasing and decreasing a pressure of the anode gas flow path inside the cell stack as a unit period. In the fuel cell system to perform,
A determination processing procedure for determining whether the electrolyte membrane is in a dry state or a wet state during the operation;
The pressure increase ratio changing processing procedure for making the rate of pressure increase per unit cycle larger when the electrolyte membrane is in a wet state when the electrolyte membrane is in a dry state based on the determination result. A method for operating the fuel cell system according to claim 1.

Images