JP2006351318A - Fuel cell system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology capable of suppressing local fluctuation in supply volume of fuel gas inside the fuel cell at the start of fuel gas supply. <P>SOLUTION: A process of circulating again the exhaust gas from a plurality of internal fuel gas passages installed in the fuel cell into the internal fuel gas passages is started, and supply of a fuel gas to the plurality of internal fuel gas passages is started at a timing after a prescribed circulation time or more larger than zero from the start of circulation has elapsed. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique for supplying fuel gas to a fuel cell.

  In recent years, attention has been focused on fuel cells that generate power using fuel gas. When such a fuel cell is started, fuel gas is supplied to the fuel cell, and then power generation is started (see, for example, Patent Document 1).

JP 2002-216812 A

  By the way, when the fuel cell is started, components other than the fuel component (for example, hydrogen) used for the electrochemical reaction (hereinafter also referred to as “impurities”, for example, nitrogen and water) accumulate in the fuel gas flow path inside the fuel cell. There may be. Then, at the beginning of the fuel gas supply, the local variation of the fuel gas supply amount inside the fuel cell becomes large, and a problem may occur.

  For example, some fuel cells have a stack structure in which a plurality of cells are stacked. A “cell” is a structure having an electrolyte layer, and an anode and a cathode sandwiching the electrolyte layer. At the beginning of the fuel gas supply, the fuel gas flows while extruding impurities accumulated in the gas flow path of each cell. Here, the fuel gas may start to flow in order from the cell where the gas easily flows. For example, when a plurality of cells include cells that are relatively close to and far from the fuel gas supply path connected to the fuel cell, the fuel gas is a cell that is relatively close to the fuel gas supply path. May flow in order. Then, the fuel gas supplied to the fuel cell passes through a relatively close cell, and impurities remain in a relatively far cell. Here, when exhaust gas is discharged outside to discharge impurities, a large amount of fuel gas may be discharged. In addition, the amount of fuel gas supplied to some cells is insufficient, and power generation may not be started.

  Such a problem is not limited to fuel cells having a stack structure, but is common to fuel cells having a plurality of internal fuel gas flow paths. For example, it is a problem common to fuel cells having a structure in which the flow path of the fuel gas branches into a plurality of tributaries inside the fuel cell.

  The present invention has been made to solve the above-described conventional problems, and is a technique capable of suppressing the local variation in the amount of fuel gas supplied inside the fuel cell at the beginning of the fuel gas supply. The purpose is to provide.

  In order to solve at least a part of the above problems, a fuel cell system according to the present invention includes a fuel cell having a plurality of internal fuel gas passages, and a fuel gas supply for supplying fuel gas to the plurality of internal fuel gas passages. A circulation pump that circulates exhaust gas from the plurality of internal fuel gas passages again to the plurality of internal fuel gas passages, and a control portion that controls the respective portions, the control portion including the fuel As the supply mode for supplying the fuel gas when the battery system is activated, the fuel gas supply unit starts driving the circulation pump and at a timing when a predetermined circulation time greater than zero has elapsed from the start of driving the circulation pump. A first supply mode for starting the supply of the fuel gas.

  According to this fuel cell system, since the supply of fuel gas is started at a timing when a predetermined circulation time greater than zero has elapsed since the start of exhaust gas circulation, the supplied fuel gas has a plurality of internal fuel gas flow paths. Is dispersed in a plurality of internal fuel gas flow paths by the gas circulating through the gas. As a result, it is possible to suppress an increase in local variation in the fuel gas supply amount inside the fuel cell at the beginning of the fuel gas supply.

  In the fuel cell system, the control unit further starts driving the circulation pump as the supply mode at the time of starting the fuel cell system, and from the supply start timing of the fuel gas in the first supply mode. A second supply mode in which the fuel gas supply unit starts supplying the fuel gas at an earlier timing, and the fuel cell system further correlates with the amount of impurities in the plurality of internal fuel gas flow paths. An impurity amount parameter measuring unit that measures a certain impurity amount parameter value, wherein the control unit measures the impurity amount parameter value by using the impurity amount parameter measuring unit at the start-up, and (i) the impurity amount parameter If the value indicates that the amount of impurities is relatively large, the fuel is in accordance with the first supply mode. Performs the supply of the scan, (ii) the amount of impurities parameter values, to indicate that the amount of the impurity is relatively small, it is preferable to perform the supply of the fuel gas in accordance with the second supply mode.

  According to this configuration, when the impurity amount parameter value indicates that the impurity amount is relatively large, the fuel gas supply start timing is set to a relatively late timing according to the first supply mode, and the impurity amount parameter is set. When the value indicates that the amount of impurities is relatively small, the fuel gas supply start timing is set to a relatively early timing according to the second supply mode. As a result, while suppressing the time required for starting up the fuel cell system from becoming excessively long, it is possible to suppress an increase in local variation of the fuel gas supply amount inside the fuel cell at the beginning of the fuel gas supply. It becomes possible.

  In each of the fuel cell systems described above, exhaust gas exhaust passages from the plurality of internal fuel gas passages extending from the plurality of internal fuel gas passages to the outside, and openable / closable provided in the exhaust passages A discharge valve, and when the fuel cell system is activated, the control unit opens the discharge valve after the start of the fuel gas supply in accordance with the first supply mode, and thereby releases the exhaust gas to the outside. It is good also as performing the discharge process to discharge.

  According to this configuration, at the beginning of the fuel gas supply, it is suppressed that the local variation in the fuel gas supply amount inside the fuel cell is increased, so that an excessive amount of exhaust gas from the internal fuel gas flow path is excessive. Including fuel gas is suppressed. As a result, impurities can be discharged to the outside while suppressing an excessive amount of fuel gas from being discharged to the outside.

  In the fuel cell system, the control unit stops the supply of the fuel gas to the fuel gas supply unit before the discharge process, and executes the discharge process in a state where the supply of the fuel gas is stopped. Is preferred.

  According to this structure, it can suppress that the supplied fuel gas is immediately discharged | emitted outside.

  Each fuel cell system further includes a differential pressure parameter measurement unit that measures a differential pressure parameter value correlated with a differential pressure between an upstream pressure and a downstream pressure of the exhaust valve, and the control unit includes: In the discharge process, the differential pressure parameter value is measured using the differential pressure parameter measurement unit, and the differential pressure parameter value is a value indicating that the differential pressure is equal to or less than a predetermined allowable differential pressure. The open state of the discharge valve may be continued.

  According to this configuration, a large amount of impurities can be discharged to the outside as compared with the case where the discharge valve is closed in a state where the differential pressure is larger than the allowable differential pressure.

  The present invention can be realized in various forms, for example, a fuel cell system, a starting method or apparatus for the fuel cell system, a computer program for realizing the functions of these methods or apparatuses, and the computer The present invention can be realized in the form of a recording medium on which the program is recorded, a data signal that includes the computer program and embodied in a carrier wave, a vehicle on which the fuel cell system is mounted as a driving power source, and the like.

Next, embodiments of the present invention will be described in the following order based on examples.
A. First embodiment:
B. Second embodiment:
C. Third embodiment:
D. Variations:

A. First embodiment:
FIG. 1 is an explanatory diagram showing the configuration of a fuel cell system as an embodiment of the present invention. The fuel cell system 900 includes a fuel cell stack 100, a hydrogen supply system 200, a circulation pump 400, an exhaust valve 500, and a control unit 600.

  The fuel cell stack 100 (hereinafter also simply referred to as “stack 100”) is a solid polymer electrolyte type fuel cell having a stack structure in which a plurality of fuel cells (not shown) (also referred to as “single cells”) are stacked. Each single cell (not shown) includes an internal anode flow channel 102 and an internal cathode flow channel 104 inside (in FIG. 1, the internal anode flow channel 102 and the internal cathode flow channel 104 of one single cell are provided. Is shown as a representative). Hydrogen gas as a fuel gas is supplied to the internal anode channel 102 (described later). Air as an oxidizing gas containing oxygen is supplied to the internal cathode channel 104. The fuel cell stack 100 generates power by an electrochemical reaction using hydrogen flowing through the internal anode flow channel 102 and oxygen flowing through the internal cathode flow channel 104. In FIG. 1, the gas flow path on the cathode side is not shown.

  The hydrogen supply system 200 includes a hydrogen tank 210 that stores hydrogen, a supply valve 220, and a pressure regulating valve 230. The hydrogen tank 210 is connected to the internal anode flow path 102 of the stack 100 via the hydrogen supply path 300. The hydrogen supply path 300 is provided with a supply valve 220 and a pressure regulating valve 230 in order from the side close to the hydrogen tank 210. When the supply valve 220 is in the open state, the hydrogen gas in the hydrogen tank 210 is supplied to the stack 100. When the supply valve 220 is in the closed state, the supply of hydrogen gas is shut off. The pressure regulating valve 230 reduces the pressure of the hydrogen gas from the hydrogen tank 210 to a predetermined pressure. The setting of the pressure regulating valve 230, that is, the pressure after the pressure reduction can be appropriately set according to the size of a load (not shown) connected to the stack 100.

  An anode off gas channel 310 is connected to the downstream side of the internal anode channel 102. Exhaust gas from the internal anode channel 102 (hereinafter also referred to as “anode off gas”) is discharged from the stack 100 via the anode off gas channel 310. The downstream side of the anode off-gas flow path 310 is connected to the hydrogen supply path 300 between the pressure regulating valve 230 and the stack 100 (confluence point C). As described above, the entire downstream side of the junction C of the hydrogen supply path 300 and the anode off-gas flow path 310 form a circulation flow path for circulating the anode off-gas to the internal anode flow path 102 again. Yes.

  A circulation pump 400 is provided in the middle of the anode off gas flow path 310. The circulation pump 400 forcibly circulates the anode off gas. During the power generation operation, the circulation pump 400 is driven, and the anode off gas is supplied to the internal anode channel 102 again. The circulation pump 400 is also driven when the fuel cell system 900 is activated (described later).

  An offgas discharge path 320 is connected to the anode offgas flow path 310 between the downstream side of the internal anode flow path 102 and the circulation pump 400. The downstream end of the off-gas discharge path 320 is open to the atmosphere. The off-gas discharge path 320 is for discharging the gas flowing through the anode off-gas flow path 310, that is, the anode off-gas, to the outside of the circulation path (the anode off-gas flow path 310 and the hydrogen supply path 300) to the internal anode flow path 102. It is a gas flow path. Further, an exhaust valve 500 is provided in the off-gas discharge path 320.

  Incidentally, the anode off gas may contain impurities. Examples of the impurities include moisture generated by an electrochemical reaction, nitrogen that has permeated through an electrolyte layer (not shown) from the internal cathode channel 104 and reached the internal anode channel 102, moisture, and the like. In the present specification, the impurities in the internal anode channel 102 mean components other than the fuel component (for example, hydrogen) used in the electrochemical reaction. Such impurities remain without being consumed. Therefore, the amount of impurities in the anode off gas gradually increases. Here, when the exhaust valve 500 is opened, a gas containing impurities is exhausted from the anode off-gas flow path 310 to the outside. As a result, the amount of impurities in the gas flowing through the internal anode channel 102 can be reduced. The controller 600 prevents the amount of impurities from excessively increasing by opening and closing the exhaust valve 500 intermittently. Hereinafter, the discharge process of opening the exhaust valve 500 and discharging the gas is also referred to as “purge process”.

  The control unit 600 controls the overall operation state of the fuel cell system 900 by receiving a data signal from each component constituting the fuel cell system 900 and outputting a drive signal to each component (described later). . The control unit 600 includes a CPU and a memory, and implements various functions by executing computer programs. Such a computer program can be supplied in a form recorded on a computer-readable recording medium. Further, part or all of the functions of the control unit 600 may be realized by hardware. In addition, a timer 610 is connected to the control unit 600. The controller 600 can measure various times by using the timer 610 (described later).

  FIG. 2 is an explanatory diagram showing an outline of a comparative example of the startup process of the fuel cell system 900. The state of the fuel cell system 900 changes in the order of (A), (B), (C), and (D) in FIG. Hereinafter, the part (A) of FIG. 2 is referred to as “FIG. 2 (A)”. The same applies to the other parts (B) to (D) and the parts (A) to (D) of FIG.

  In the fuel cell stack 100 in each of the parts (A) to (D) of FIG. 2, the internal anode flow paths 102a, 102b, and 102c of three single cells among the plurality of single cells are representatively shown. . The upstream sides of the flow paths 102 a, 102 b, 102 c merge and are connected to the hydrogen supply path 300. Here, the first internal anode flow path 102a and the third internal anode flow path 102c are disposed at a position farther from the hydrogen supply path 300 than the second internal anode flow path 102b. On the other hand, the downstream sides of the respective flow paths 102 a, 102 b, 102 c merge and are connected to the anode offgas flow path 310. Further, the first internal anode flow channel 102a and the third internal anode flow channel 102c are disposed at a position farther from the anode off-gas flow channel 310 than the second internal anode flow channel 102b.

  FIG. 2 (A) shows the fuel cell system 900 before starting, that is, in a stopped state. In this stop state, the supply valve 220 and the exhaust valve 500 are closed. Further, impurities are accumulated in the internal anode channel 102. Such impurities include, for example, nitrogen that has passed through the electrolyte (not shown) from the internal cathode channel (not shown) and moved to the internal anode channel 102.

  The hydrogen supply path 300 between the hydrogen tank 210 and the supply valve 220 is hatched. This hatching indicates that the hydrogen concentration of the gas in the channel portion is higher than the hydrogen concentration of the gas in the channel portion without hatching. In other words, the impurity concentration of the gas in the channel portion with hatching is lower than the impurity concentration of the gas in the channel portion without hatching. The same applies to other parts of FIG. 2 and FIG. 3 described later.

  In the comparative example of the startup process, first, the control unit 600 (FIG. 1) opens the supply valve 220 and supplies hydrogen to the stack 100. Then, the fuel cell system 900 is in a state shown in FIG. Hydrogen is supplied to the stack 100 via the hydrogen supply path 300.

  Next, the control unit 600 drives the circulation pump 400. Then, the fuel cell system 900 is in the state shown in FIG. The circulation pump 400 forcibly circulates the anode off gas from the stack 100.

  In the example of FIG. 2C, the hydrogen supplied to the stack 100 passes through the second internal anode channel 102b and is discharged to the anode off-gas channel 310. On the other hand, impurities remain in the other internal anode channels 102a and 102c. The reason is as follows. The internal anode channels 102a and 102c are farther from the hydrogen supply channel 300 and the anode offgas channel 310 than the second internal anode channel 102b. In other words, the length of the flow path from the hydrogen supply path 300 to the anode off-gas flow path 310 is longer when passing through the other internal anode flow paths 102a and 102c than when passing through the second internal anode flow path 102b. Is long. Therefore, hydrogen tends to push out impurities in the flow path when flowing through the second internal anode flow path 102b than when flowing through the other internal anode flow paths 102a and 102c. Then, hydrogen first passes through the second internal anode flow path 102b, and hydrogen flows through the other internal anode flow paths 102a and 102c with a delay. That is, at the beginning of the circulation, impurities remain in the internal anode channels 102a and 102c.

  Next, the controller 600 opens the exhaust valve 500 to discharge impurities. Then, the fuel cell system 900 is in the state shown in FIG. In the example shown in FIG. 2D, impurities remain in the internal anode channels 102a and 102c. As a result, gas with a high hydrogen concentration may be discharged from the off-gas discharge path 320.

  Thereafter, the control unit 600 repeatedly executes the process of opening the exhaust valve 500 for a predetermined time at a predetermined timing. Then, the impurities are discharged, and instead, the fuel gas is supplied, so that the hydrogen concentration of the gas circulating in the internal anode channel 102 increases. Then, the control unit 600 causes the stack 100 to start power generation. If impurities remain in the internal anode channel of a single cell, power generation may not be started in that single cell. Therefore, in order to quickly supply hydrogen to all the single cells, the set pressure of the pressure regulating valve 230 at the time of startup (that is, the pressure of the fuel gas supplied to the stack 100) is set to be higher than the pressure during normal power generation operation. May be set to a higher value.

  FIG. 3 is an explanatory diagram showing an overview of the startup process of the fuel cell system 900 in the first embodiment. FIG. 4 is a flowchart showing the procedure of the startup process of the fuel cell system 900 in the first embodiment. In this activation process, the state of the fuel cell system 900 changes in the order of (A), (B), (C), and (D) in FIG. Each part (A) to (D) of FIG. 3 shows a fuel cell system 900 similar to that of FIG. Moreover, the meaning of the hatching attached to the flow path is the same as the hatching in FIG.

  In the first step S100 (FIG. 4), the control unit 600 (FIG. 1) first starts driving the circulation pump 400. Then, the gas (impurities) accumulated in the internal anode channel 102, the anode off gas channel 310, and the hydrogen supply channel 300 starts to circulate.

  In the next step S110, the control unit 600 measures the operation time of the circulation pump 400 and waits until the operation time reaches a predetermined time or more. Then, the fuel cell system 900 is in the state shown in FIG. At this stage, the supply valve 220 and the exhaust valve 500 remain closed. Further, gas (impurities) circulates in any of the internal anode flow paths 102a, 102b, 102c. When the circulation pump 400 starts driving, the gas (impurities) supplied by the stack 100 by the circulation pump 400 first passes through the second internal anode channel 102b, as in the example shown in FIG. Then, as time passes, the gas accumulated in the other internal anode channels 102a and 102c also starts to flow. After the elapse of the predetermined time, the gas circulates in any of the internal anode channels 102a, 102b, 102c. Here, the predetermined time can be determined in advance based on experiments (for example, about 3 seconds) so as to be sufficient time for the gas to circulate through the internal anode channel of each cell.

  In the next step S120, the control unit 600 opens the supply valve 220. Hydrogen is supplied to the stack 100 via the hydrogen supply path 300. Then, the fuel cell system 900 is in a state shown in FIG. At this time, hydrogen is mixed with the gas (impurities) supplied by the circulation pump 400 and then supplied to the stack 100. In the hydrogen supply path 300 downstream from the junction C, the hydrogen concentration is lowered by mixing with impurities. Therefore, the hatching attached to the hydrogen supply path 300 downstream from the junction C is thinner than the upstream side.

  The hydrogen supplied to the stack 100 is supplied to the internal anode channel of each cell. In the first embodiment, hydrogen is supplied after the operation time of the circulation pump 400 reaches a predetermined time or more, that is, after the gas circulates through the internal anode flow path of each cell. Therefore, the gas supplied to the stack 100 can smoothly pass through the internal anode channel of each cell. As a result, hydrogen is supplied to the internal anode flow paths of each cell without being biased to some internal anode flow paths as in the comparative example shown in FIG. Then, the fuel cell system 900 is in a state shown in FIG. In the example of FIG. 3C, hydrogen passes through any of the three internal anode channels 102a, 102b, 102c. Impurities are discharged from any of the internal anode channels 102a, 102b, 102c. Thus, in the first embodiment, compared to the comparative example, it is possible to reduce the amount of impurities remaining in the stack 100 without being discharged at the beginning of hydrogen supply. As a result, the hydrogen concentration of the gas circulating in the stack 100 is lower than that in the comparative example.

  In the next step S130, the control unit 600 starts the discharge process. In the first embodiment, the control unit 600 repeatedly executes a process of closing the exhaust valve 500 after opening the exhaust valve 500 for a predetermined time (hereinafter also referred to as “predetermined time purge process”) at a predetermined timing. When the control unit 600 opens the exhaust valve 500, the fuel cell system 900 is in the state shown in FIG. In the first embodiment shown in FIG. 3D, a gas having a high impurity concentration and a low hydrogen concentration is discharged as compared with the comparative example shown in FIG. When impurities are discharged, fuel gas is supplied instead, so that the hydrogen concentration of the gas circulating in the internal anode channel 102 increases.

  In the next step S140, the control unit 600 causes the stack 100 to start power generation. For example, the control unit 600 switches a switch (not shown) that can electrically connect the stack 100 and a load (not shown) to “on” and supplies generated power to the load.

  As described above, in the first embodiment, since the hydrogen gas is supplied after the circulation pump 400 is driven for a predetermined time larger than zero in advance, the hydrogen gas is supplied inside the fuel cell at the beginning of the hydrogen gas supply. An increase in local variation in amount can be suppressed. As a result, when the exhaust valve 500 is opened, it is possible to suppress an excessive increase in the hydrogen concentration of the gas discharged to the outside. That is, it is possible to prevent an excessive amount of hydrogen discharged without contributing to power generation. Moreover, since it is suppressed that the amount of hydrogen supplied to one part cell decreases excessively, it can also suppress that it becomes impossible to start electric power generation in one part cell. As a result, the fuel cell system 900 can be activated without excessively increasing the pressure of the fuel gas supplied to the stack 100.

  Note that power generation is started regardless of the hydrogen concentration in the internal anode channel 102 after the hydrogen supply (FIG. 4: step S120), immediately before the startup process (that is, at the start of the startup process). If it is sufficiently high, the discharge process (step S130) may be omitted.

B. Second embodiment:
FIG. 5 is a flowchart showing the procedure of the startup process of the fuel cell system 900 in the second embodiment. There are two differences from the activation process of the first embodiment shown in FIG. The first difference is that the controller 600 executes the discharge process with the supply valve 220 closed. The second difference is that the time during which the control unit 600 (FIG. 1) opens the exhaust valve 500 (hereinafter also referred to as “opening time”) in the discharge process is set to a long time that can sufficiently discharge the anode off-gas. It is a point. The configuration of the fuel cell system is the same as that of the fuel cell system 900 (FIG. 1) of the first embodiment.

  The processes in steps S200 to S220 are the same as steps S100 to S120 in FIG.

  When hydrogen is supplied to the stack 100 (FIG. 1) (S220), the control unit 600 closes the supply valve 220 in the next step S230, and opens the exhaust valve 500 in the next step S240. Then, in the next step S245, the control unit 600 measures an elapsed time after opening the exhaust valve 500. This elapsed time is a time during which the exhaust valve 500 is kept open (open time). The control unit 600 waits for the opening time to be equal to or longer than a predetermined discharge time.

  In the second embodiment, the predetermined discharge time is set in advance to a long time that can sufficiently discharge the anode off gas. On the other hand, the pressure on the downstream side of the exhaust valve 500 in the off-gas discharge path 320 (FIG. 1) is the same as the atmospheric pressure (about 100 kPa). In steps S240 and S245, the anode off-gas is discharged until the pressure on the upstream side of the exhaust valve 500 in the off-gas discharge path 320 is substantially the same as the pressure on the downstream side (atmospheric pressure).

  When the anode off gas is discharged to the outside, the controller 600 (FIG. 1) closes the exhaust valve 500 in the next step S250, and opens the supply valve 220 in the next step S260. Then, hydrogen is supplied to the stack 100. In the next step S270, the control unit 600 causes the stack 100 to start power generation.

  As described above, in the startup process of the second embodiment, the control unit 600 executes the discharge process with the supply valve 220 closed (that is, with the supply of fuel gas stopped). As a result, the supplied hydrogen can be prevented from being immediately discharged to the outside. Furthermore, in the second embodiment, the control unit 600 closes the supply valve 220 while the exhaust valve 500 is open. As a result, it is possible to further suppress the supplied hydrogen from being immediately discharged to the outside. However, the supply valve 220 may be opened before the exhaust valve 500 is closed.

  In the discharge process of the second embodiment, the open time is set to a long time (predetermined discharge time) that can sufficiently discharge the anode off gas. In particular, in the discharge process of the second embodiment, the predetermined discharge time is determined by the difference between the pressure on the upstream side and the pressure on the downstream side of the exhaust valve 500 after the exhaust valve 500 (FIG. 1) is opened. It is set to the time until the pressure becomes less than the allowable differential pressure. That is, the exhaust valve 500 is opened until the differential pressure becomes equal to or lower than the allowable differential pressure (hereinafter, the discharge process in which the open time is longer than the discharge time is also referred to as “low differential pressure discharge process”). As a result, a larger amount of impurities can be discharged to the outside as compared with the discharge process (hereinafter referred to as “high differential pressure discharge process”) in which the exhaust valve 500 is closed in a state where the differential pressure is higher than the allowable differential pressure. Therefore, in the second embodiment, compared with the case where the high differential pressure discharge process and the supply of hydrogen are repeated, the amount of hydrogen discharged outside without contributing to power generation is prevented from being excessively increased. It becomes possible to increase the hydrogen concentration of the gas circulating in the internal anode channel 102. In this embodiment, the opening time corresponds to the “differential pressure parameter value”. The entire control unit 600 and timer 610 correspond to a “differential pressure parameter measurement unit”.

  Here, it is preferable that the allowable pressure difference is smaller than the pressure on the downstream side of the exhaust valve 500. For example, a value of 1/10 or less of the downstream pressure can be adopted (for example, 10 kPa). The time until the differential pressure becomes equal to or less than the allowable differential pressure (predetermined discharge time) can be experimentally determined in advance.

  The parameter value used for determining the timing for ending the exhaust process (timing for closing the exhaust valve 500) is not limited to the opening time, but is correlated with the differential pressure between the pressure on the upstream side and the pressure on the downstream side of the exhaust valve 500. It is possible to adopt various parameter values (differential pressure parameter values). For example, a pressure sensor can be provided on each of the upstream side and the downstream side of the exhaust valve 500 of the off gas discharge path 320. In this case, a differential pressure obtained from the measured values of these pressure sensors can be employed. In any case, in the discharge process, the control unit 600 measures the differential pressure parameter value using the differential pressure parameter measurement unit, and the differential pressure parameter value indicates that the differential pressure is equal to or less than a predetermined allowable differential pressure. It is preferable to keep the exhaust valve 500 open until the value is reached.

  In the activation process shown in FIG. 5, the opening time may be shorter than the discharging time as described above. In any case, it is preferable that the control unit 600 executes the discharge process in a state in which the supply of the fuel gas is stopped regardless of the length of the open time in the discharge process.

  On the other hand, the control unit 600 may perform the discharge process while supplying the fuel gas. For example, step S230 may be omitted in the startup process shown in FIG. In this case, it is preferable to set the fuel gas supply speed (supply amount per unit time) to a small value so that the pressure on the upstream side of the exhaust valve 500 is sufficiently lowered by the discharge process. For example, the supply speed can be reduced by reducing the opening degree of the supply valve 220.

  Incidentally, the control unit 600 may determine whether or not to execute the discharge process (step S240) according to the hydrogen concentration in the internal anode channel 102. For example, it is possible to employ a configuration in which the control unit 600 repeatedly executes the processes of steps S230 to S260 until the hydrogen concentration becomes a predetermined threshold value or more.

  Various methods can be adopted as a method for obtaining the hydrogen concentration. For example, the fuel cell system 900 (for example, the hydrogen supply path 300 or the anode off-gas flow path 310) can be provided with a hydrogen concentration sensor that measures the hydrogen concentration of the gas circulating in the internal anode flow path 102. The controller 600 can employ the measurement value of the hydrogen concentration sensor as the hydrogen concentration.

  Moreover, you may employ | adopt the method of using the estimated value of hydrogen concentration instead of a measured value. FIG. 6 shows the elapsed time since the last stop of the operation of the fuel cell system 900 (hereinafter also referred to as “stop time”), the hydrogen partial pressure in the internal anode channel 102, and the internal pressure in the internal anode channel 102. It is a graph which shows an example of correspondence (it is also called "time partial pressure relationship" hereafter) with nitrogen partial pressure. The horizontal axis represents elapsed time (stop time), and the vertical axis represents partial pressure. The solid line graph indicates the hydrogen partial pressure, and the broken line graph indicates the nitrogen partial pressure.

  When the operation of the fuel cell system 900 is stopped, hydrogen in the internal anode channel 102 passes through the electrolyte layer of the stack 100 and moves to the internal cathode channel 104, so that the hydrogen partial pressure in the internal anode channel 102 is reduced. Gradually lower. Furthermore, since nitrogen in the internal cathode channel 104 passes through the electrolyte layer and moves to the internal anode channel 102, the partial pressure of nitrogen in the internal anode channel 102 gradually increases. Such a time partial pressure relationship can be experimentally obtained in advance. In addition, the experimentally set time division relationship can be stored in advance in a memory (not shown) of the control unit 600. In this way, the control unit 600 can estimate the hydrogen partial pressure and the nitrogen partial pressure at the start of the startup process by referring to the time partial pressure relationship. Further, the control unit 600 can estimate the hydrogen concentration in the startup process by using the set pressure of the pressure regulating valve 230 in addition to the hydrogen partial pressure estimated value and the nitrogen partial pressure estimated value thus obtained.

  FIG. 7 is a schematic diagram illustrating how the control unit 600 estimates the hydrogen concentration in the startup process. In the example of FIG. 7, the ratio of the hydrogen partial pressure to the total pressure in the internal anode channel 102 (hereinafter referred to as “hydrogen partial pressure ratio”) is used as the hydrogen concentration. Further, power generation is started when the hydrogen partial pressure ratio is a predetermined ratio (for example, 70%) or more. In addition, the set pressure of the pressure regulating valve 230 in the startup process (that is, the total pressure of the internal anode channel 102 after the supply of hydrogen) is 200 kPa.

  In the example of FIG. 7, the estimated hydrogen partial pressure is 0 kPa and the nitrogen partial pressure estimated value is 100 kPa at the start of the startup process. These estimated values are obtained from the time partial pressure relationship (FIG. 6) and the measured stop time value. The partial pressure of other components (for example, oxygen) will be omitted because it is sufficiently small. When the supply valve 220 is opened in this state (FIG. 5: Step S220), the stack 100 (inner anode flow path 102) is pressurized with hydrogen, and the total pressure becomes 200 kPa. Here, since the nitrogen partial pressure does not change, the controller 600 can estimate that the remainder (100 kPa) obtained by subtracting the nitrogen partial pressure (100 kPa) from the total pressure (200 kPa) is the hydrogen partial pressure. Further, the control unit 600 can estimate that the hydrogen partial pressure ratio is 50%. Since the hydrogen partial pressure ratio estimated value (50%) is less than the predetermined ratio (70%), the control unit 600 closes the supply valve 220 (S230) and executes the discharge process (S240).

  When the discharge process is executed, the total pressure becomes almost atmospheric pressure (100 kPa). Here, the ratio between the hydrogen partial pressure and the nitrogen partial pressure is constant. Therefore, the control unit 600 can estimate that the hydrogen partial pressure after discharge processing is 50 kPa and the nitrogen partial pressure is 50 kPa.

  Next, the control unit 600 closes the exhaust valve 500 (step S250) and supplies hydrogen again (step S260). Then, the total pressure becomes 200 kPa. Here, since the nitrogen partial pressure does not change, the control unit 600 can estimate that the remainder (150 kPa) obtained by subtracting the nitrogen partial pressure (50 kPa) from the total pressure (200 kPa) is the hydrogen partial pressure. Further, the control unit 600 can estimate that the hydrogen partial pressure ratio is 75%. Since this hydrogen partial pressure ratio estimated value (75%) is not less than the predetermined ratio (70%), the control unit 600 starts power generation (step S270).

  If the hydrogen partial pressure ratio estimated value is less than the predetermined ratio after step S260, control unit 600 repeatedly executes the processes of steps S230 to S260 until the hydrogen partial pressure ratio estimated value becomes equal to or greater than the predetermined ratio. To do. If the hydrogen partial pressure ratio estimated value is equal to or greater than the predetermined ratio after step S220, control unit 600 skips steps S230 to S260 and proceeds to step S270. As such a case, for example, the fuel cell system 900 may be restarted again immediately after the operation of the fuel cell system 900 is stopped.

  In the examples shown in FIGS. 6 and 7, the control unit 600 estimates the hydrogen concentration using the estimated values of the hydrogen partial pressure and the nitrogen partial pressure, but further, other components of impurities (for example, The hydrogen concentration may be estimated in consideration of the partial pressure of oxygen or water.

C. Third embodiment:
FIG. 8 is a flowchart showing the procedure of the startup process of the fuel cell system 900 according to the third embodiment. The difference between the first embodiment shown in FIG. 4 and the second embodiment shown in FIG. 5 is that the control unit 600 estimates the nitrogen partial pressure at the start of the start-up process, and according to the estimated value, the supply valve This is that the timing for opening 220 is changed. The configuration of the fuel cell system is the same as that of the fuel cell system 900 (FIG. 1) of the first embodiment.

  In the first step S300, the controller 600 estimates the nitrogen partial pressure in the internal anode channel 102. As a method for estimating the nitrogen partial pressure, various methods can be employed. For example, it is possible to store a time division relationship (FIG. 6) set experimentally in advance in a memory (not shown) of the control unit 600. The controller 600 can estimate the nitrogen partial pressure by measuring the stop time (the elapsed time since the last stop of the operation of the fuel cell system 900) and referring to this time-partial pressure relationship. In this case, the stop time corresponds to the “impurity amount parameter value”. The entire control unit 600 and timer 610 correspond to an “impurity amount parameter measurement unit”.

  Further, the fuel cell system 900 can be provided with a pressure sensor that measures the pressure in the internal anode flow channel 102 and a hydrogen concentration sensor that measures the hydrogen concentration of the gas in the internal anode flow channel 102. The controller 600 can estimate the hydrogen partial pressure from the measured values of these sensors. Further, it is possible to estimate that the remainder obtained by subtracting the hydrogen partial pressure estimated value from the measured value (total pressure) of the pressure sensor is the nitrogen partial pressure. Note that such a hydrogen concentration sensor is, for example, a circulation flow path from the internal anode flow path 102 to the internal anode flow path 102 again (on the downstream side of the junction C of the hydrogen supply path 300 and the anode off-gas flow path 310). Can be provided. The same applies to the pressure sensor. In this case, the entire hydrogen concentration and pressure correspond to the “impurity amount parameter value”. The whole of the hydrogen concentration sensor and the pressure sensor corresponds to an “impurity amount parameter measuring unit”.

  In next step S310, control unit 600 determines whether or not the estimated value of the nitrogen partial pressure is equal to or higher than a predetermined partial pressure. If the nitrogen partial pressure is greater than or equal to the predetermined partial pressure, control unit 600 proceeds to step S320. In step S320, the control unit 600 executes the same startup process as that of the first embodiment shown in FIG. Here, the controller 600 uses the first predetermined time as the predetermined time in step S110 (FIG. 4). The first predetermined time is a value larger than zero, and is a value larger than a second predetermined time described later.

  On the other hand, if the nitrogen partial pressure is less than the predetermined partial pressure, control unit 600 proceeds to step S330. In step S330, similarly to step 320, the control unit 600 executes the same startup process as in the first embodiment shown in FIG. Here, the controller 600 uses the second predetermined time as the predetermined time in step S110 (FIG. 4). This second predetermined time is a value smaller than the first predetermined time used in step S320.

  As described above, in the third embodiment, when the nitrogen partial pressure estimated value is relatively high, the control unit 600 sets the drive time of the circulation pump 400 until hydrogen is supplied to a relatively long first predetermined time. (Step S320). The higher the nitrogen partial pressure, the longer the time required to push out nitrogen in the internal anode channel 102. Therefore, the controller 600 sets the drive time of the circulation pump 400 before the start of hydrogen supply to a relatively long time (first predetermined time), so that the hydrogen supply amount in the fuel cell is locally increased at the beginning of hydrogen supply. It is possible to suppress an increase in general variation.

  On the other hand, when the nitrogen partial pressure estimated value is relatively low, control unit 600 sets the drive time of circulation pump 400 until hydrogen is supplied to a relatively short second predetermined time (step S330). The lower the nitrogen partial pressure, the shorter the time required to push out nitrogen in the internal anode channel 102. Therefore, the control unit 600 can prevent the startup process from taking excessive time by setting the drive time of the circulation pump 400 before the start of hydrogen supply to a relatively short time (second predetermined time).

  The predetermined partial pressure, the first predetermined time, and the second predetermined time are set to values that can suppress an increase in local variation in the amount of hydrogen supplied inside the fuel cell at the beginning of hydrogen supply. It may be set in advance based on experiments.

  Moreover, the starting process which the control part 600 performs by step S320, S330 is not restricted to the process shown in FIG. 4, A various process is employable. For example, you may employ | adopt the process which abbreviate | omitted step S130 from the procedure of FIG. Moreover, you may employ | adopt the same starting process as 2nd Example shown in FIG. In this case, the control unit 600 may use the first predetermined time and the second predetermined time in accordance with the nitrogen partial pressure estimated value in step S210 (FIG. 5). Here, as the processes after step S220, various processes described in the second embodiment can be employed.

  The start-up process executed in step S320 includes driving the circulation pump 400 and starting the supply of hydrogen after operating the circulation pump 400 for a predetermined time (first predetermined time) greater than zero (supply valve). Any starting process including a process (for example, the process of steps S100 to S120 in FIG. 4) can be employed.

  The start-up process executed in step S330 is a process of starting the supply of hydrogen (opening the supply valve 220) after operating the circulation pump 400 for a predetermined time (second predetermined time) smaller than the first predetermined time. Any start-up process including can be adopted. Furthermore, the start-up process executed in step S330 is not limited to the start-up process for driving the circulation pump 400 before the start of hydrogen supply, and the start-up process for starting the hydrogen supply simultaneously with the start of driving of the circulation pump 400 may be adopted. Good. Moreover, you may employ | adopt the starting process which starts the drive of the circulation pump 400 after starting hydrogen supply. In general, as the start-up process executed in step S330, any start-up process in which the circulation pump 400 is driven and the hydrogen supply start timing is earlier than the timing in the start-up process in step S320 can be employed.

D. Variations:
In addition, elements other than the elements claimed in the independent claims among the constituent elements in the above embodiments are additional elements and can be omitted as appropriate. The present invention is not limited to the above-described examples and embodiments, and can be implemented in various modes without departing from the gist thereof. For example, the following modifications are possible.

Modification 1:
In the third embodiment shown in FIG. 8, the control unit 600 converts the stop time and the hydrogen concentration into the nitrogen partial pressure, but such a measured value (stop time and hydrogen concentration) is converted into the nitrogen partial pressure. The hydrogen supply start timing may be adjusted by using it without conversion. For example, when the stop time is used, the control unit 600 proceeds to step S320 when the stop time is equal to or greater than a predetermined threshold time, and proceeds to step S330 when the stop time is less than the threshold time. It may be migrated.

  The amount of impurities is not limited to the partial pressure of nitrogen, and various values can be adopted. For example, it is possible to use the partial pressure of impurities including components other than nitrogen (for example, oxygen and water). Further, the concentration of impurities in the internal anode channel 102 can be used instead of the partial pressure of impurities. In any case, it is preferable to use the amount of impurities containing at least the most abundant component (usually nitrogen) among the impurity components.

  The parameter values used for determining the timing for starting the hydrogen supply are not limited to the stop time and the hydrogen concentration, but various parameter values (impurity amount parameter values) correlated with the impurity amounts in the plurality of internal anode channels 102. Can be adopted. For example, the amount of impurities may increase as the integrated power generation amount increases. In such a case, the integrated power generation amount can be adopted. In the example of FIG. 8, the control unit 600 switches the timing of starting hydrogen supply to two stages, but may switch to three or more stages. Further, the control unit 600 may continuously change the hydrogen supply start timing according to the impurity amount parameter value. In any case, the control unit 600 measures the impurity amount parameter value using the impurity amount parameter measurement unit at the time of startup, and (i) when the impurity amount parameter value indicates that the impurity amount is relatively large. The hydrogen supply start timing is set to a relatively late timing, and (ii) when the impurity amount parameter value indicates that the impurity amount is relatively small, the hydrogen supply start timing is set to a relatively early timing. It is preferable.

Modification 2:
In the example of FIG. 3, the case where the channel length from the hydrogen supply channel 300 through the internal anode channel 102 of each single cell to the anode offgas channel 310 is different for each single cell has been described. Any other configuration can be adopted. For example, the stack 100 may be configured such that the channel length from the hydrogen supply channel 300 to the anode offgas channel 310 through the internal anode channel 102 of each unit cell is the same for each unit cell. In any case, at the beginning of gas supply to the fuel cell, gas (impurities) in the internal anode flow paths 102 of some single cells may remain without being discharged due to various factors. Factors include, for example, a difference in the shape of the flow path from the hydrogen supply path 300 to each internal anode flow path 102, an individual difference in the shape of the internal anode flow path 102 in each single cell, and the like. Therefore, as in the above-described embodiments, if the supply of hydrogen gas is started after the circulation pump is driven for a predetermined time, local variation in the amount of hydrogen gas supplied inside the fuel cell at the beginning of the hydrogen gas supply. Can be prevented from becoming large.

Modification 3:
Also in the starting process of the first embodiment shown in FIG. 4, as in the second embodiment, whether or not the control unit 600 (FIG. 1) executes the discharge process (step S <b> 130) is determined. It is good also as determining according to the hydrogen concentration in. Specifically, the control unit 600 may repeatedly execute the discharge process until the hydrogen concentration becomes equal to or higher than the threshold value. Also in this case, the controller 600 can determine whether or not the hydrogen concentration is equal to or higher than a predetermined threshold value, as in the second embodiment. For example, a correspondence relationship (hereinafter also referred to as “concentration change correspondence relationship”) between the hydrogen concentration before the purge treatment for a certain time and the hydrogen concentration after the purge treatment for a certain time (after supplying hydrogen) (hereinafter also referred to as “concentration change correspondence”) is experimentally obtained in advance. It is possible to keep. Further, in addition to the time-divided pressure relationship (FIG. 6), such a concentration change correspondence relationship can also be stored in advance in a memory (not shown) of the control unit 600. In this way, the control unit 600 can estimate the hydrogen concentration while repeatedly executing the purge process for a predetermined time by referring to these correspondences. Here, the control unit 600 measures the stop time, counts (measures) the number of times the purge process is repeated, and estimates the hydrogen concentration by using the stop time and the number of times.

  Note that the controller 600 can use various parameter values (hydrogen concentration parameter values) that are correlated with the hydrogen concentration in the internal anode flow channel 102 as well as the hydrogen concentration. For example, the measurement value of the hydrogen concentration sensor, the combination of the measurement value of the stop time and the measurement value of the number of repetitions of the discharge process, the impurity amount parameter value described in the first modification, and the like can be employed. Here, the control unit 600 measures the hydrogen concentration parameter value using the hydrogen concentration parameter measurement unit before executing the discharge process, and (i) the hydrogen concentration parameter value is less than a predetermined threshold value. (Ii) If the hydrogen parameter value indicates that the hydrogen concentration is equal to or higher than a predetermined threshold value, the power generation is performed without executing the discharge process. It is preferable to start. The same applies to the startup process of the second embodiment shown in FIG.

  In addition, the control unit 600 may adjust the opening time in the purge process (for example, FIG. 4: S130) according to the hydrogen concentration. Specifically, the open time may be set to a longer value as the hydrogen concentration is lower. In this way, impurities can be discharged while preventing an excessive amount of hydrogen being discharged without contributing to power generation. Further, the controller 600 may set the open time according to the impurity concentration or partial pressure (more generally, the impurity amount parameter value correlated with the amount of impurities) instead of the hydrogen concentration. Here, when executing the discharge processing, the control unit 600 measures the impurity amount parameter value using the impurity amount parameter measurement unit, and (i) the impurity amount parameter value indicates that the impurity amount is relatively large. In this case, it is preferable to set the opening time to a relatively long time, and (ii) when the impurity amount parameter value indicates that the impurity amount is relatively small, the opening time is set to a relatively short time. . As the impurity amount parameter value used for setting the open time, the same impurity amount parameter value used for determining the hydrogen concentration and the timing of starting the hydrogen supply described in the first modification is used. Is possible. Similarly, also in the purge process of the startup process shown in FIG. 5, the control unit 600 may adjust the opening time (the discharging time in step S245) according to the impurity amount parameter value.

Modification 4:
In each of the above embodiments, as the hydrogen tank 210, various devices that store hydrogen, such as a high-pressure hydrogen tank, a liquid hydrogen tank, and a hydrogen storage alloy tank, can be used. In each of the above embodiments, the configuration of the fuel gas supply unit (in each of the above embodiments, the hydrogen supply system 200) is not limited to the configuration having the hydrogen tank 210, but is used for an electrochemical reaction for power generation. Various configurations capable of supplying a fuel gas containing a fuel component (for example, hydrogen) and controlling the start timing of the fuel gas supply can be employed. For example, instead of the hydrogen tank 210, a fuel gas supply device that has a reformer that generates hydrogen by a reforming reaction and supplies a reformed gas containing the generated hydrogen as a fuel gas may be used.

Modification 5:
In each of the above embodiments, various known pumps can be employed as the circulation pump 400. For example, a pump that performs intake and exhaust by rotational drive can be employed. Also, some pumps using rotational drive do not immediately increase to the rotational speed designated in advance (also referred to as “instructed rotational speed”) even when driving is started. Therefore, in each of the above-described embodiments, as the pump operation time to be compared with the predetermined time (for example, step S110 in FIG. 4 and step S210 in FIG. 5), the rotation speed of the pump is predetermined instead of the operation time from the start of driving. It is good also as using the driving | running time from the timing which became more than this instruction | indication rotation speed. By so doing, it takes time to increase the rotational speed of the pump, and therefore, it is possible to suppress an increase in local variation in the amount of fuel gas supplied inside the fuel cell at the beginning of fuel gas supply. The indicated rotational speed may be a value that allows the gas to circulate in the internal anode flow path of each cell, and may be determined in advance based on experiments.

Modification 6:
In each of the above embodiments, the fuel cell stack 100 having a plurality of single cells is employed as the fuel cell. However, the fuel cell is not limited to a fuel cell having a plurality of single cells, and generally a plurality of internal fuels. Any fuel cell having a gas flow path can be employed. For example, a single cell having a configuration in which the flow path of the fuel gas is branched inside the single cell may be adopted. In such a single cell, at the beginning of gas supply to the single cell, gas (impurities) in some of the tributaries may be less likely to be discharged than gases (impurities) in other tributaries. Therefore, as in the above embodiments, if the supply of the fuel gas is started after the circulation pump is driven for a predetermined time, there is a local variation in the amount of fuel gas supply inside the single cell at the beginning of the fuel gas supply. It is possible to suppress the increase. As the fuel cell, a fuel cell stack having a plurality of such single cells may be used, or a fuel cell having only one such single cell may be used.

Modification 7:
The configuration of the gas flow path on the anode side connected to the fuel cell stack 100 is not limited to the configuration of the fuel cell system 900 shown in FIG. 1, and various configurations can be adopted. For example, the downstream side of the anode off-gas channel 310 may be directly connected to the plurality of internal anode channels 102 without joining the hydrogen supply channel 300. Further, the off gas discharge channel 320 may be directly connected to the plurality of internal anode channels 102 instead of the anode off gas channel 310. In general, the anode-side gas flow path includes a plurality of anode off-gas circulation paths from a plurality of internal fuel gas flow paths to a plurality of internal fuel gas flow paths through a circulation pump, and a plurality of fuel gas supply sections. It is possible to adopt a configuration having a fuel gas supply path to the internal fuel gas flow path. Furthermore, it is preferable that the configuration of the gas flow path on the anode side includes an anode off-gas discharge flow path from the plurality of internal fuel gas flow paths to the outside of the circulation path through the discharge valve.

Modification 8:
In each of the above embodiments, the off-gas discharge path 320 may be provided with a diluting device for diluting the exhaust gas with air. In each of the above embodiments, the hydrogen concentration of the gas discharged to the outside from the off-gas discharge path 320 is suppressed from being excessively increased in the startup process. Therefore, an excessive increase in the load on the dilution device is suppressed.

Modification 9:
In each of the above-described embodiments, a solid polymer electrolyte was used as the electrolyte. In addition, various types of electrolytes such as a solid oxide electrolyte, a phosphate electrolyte, an alkaline aqueous electrolyte, and a molten carbonate electrolyte were used. Can be used.

Modification 10:
Both the above-described embodiment and modification examples relating to the activation process of FIG. 4 and the activation process of FIG. 5 can be applied to steps S320 and S330 of FIG.

BRIEF DESCRIPTION OF THE DRAWINGS Explanatory drawing which shows the structure of the fuel cell system as one Example of this invention. Explanatory drawing which shows the outline | summary of the comparative example of the starting process of the fuel cell system. Explanatory drawing which shows the outline | summary of the starting process of the fuel cell system 900 in 1st Example. The flowchart which shows the procedure of the starting process of the fuel cell system 900 in 1st Example. The flowchart which shows the procedure of the starting process of the fuel cell system 900 in 2nd Example. The graph which shows an example of the correspondence of elapsed time, hydrogen partial pressure, and nitrogen partial pressure. Schematic which shows a mode that the control part 600 estimates hydrogen concentration in a starting process. The flowchart which shows the procedure of the starting process of the fuel cell system 900 in 3rd Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Fuel cell stack 102 ... Internal anode flow path, 102a ... 1st internal anode flow path,
102b ... second internal anode flow path, 102c ... third internal anode flow path 104 ... internal cathode flow path 200 ... hydrogen supply system 210 ... hydrogen tank 220 ... supply valve 230. .. Pressure regulating valve 300 ... Hydrogen supply passage 310 ... Anode off-gas passage 320 ... Off-gas discharge passage 400 ... Circulation pump 500 ... Exhaust valve 600 ... Control unit 610 ... Timer 900 ... Fuel cell system C ... Confluence

Claims (6)

A fuel cell system,
A fuel cell having a plurality of internal fuel gas flow paths;
A fuel gas supply unit for supplying fuel gas to the plurality of internal fuel gas flow paths;
A circulation pump that circulates exhaust gas from the plurality of internal fuel gas passages again to the plurality of internal fuel gas passages;
A control unit for controlling each unit;
With
The control unit starts driving the circulation pump as a supply mode for supplying the fuel gas when starting the fuel cell system, and a predetermined circulation time greater than zero has elapsed from the start of driving the circulation pump. Having a first supply mode in which the fuel gas supply unit starts supplying the fuel gas at a timing;
Fuel cell system. The fuel cell system according to claim 1,
The control unit further starts driving the circulation pump as the supply mode at the time of startup of the fuel cell system, and at a timing earlier than the supply start timing of the fuel gas in the first supply mode. A second supply mode for causing the gas supply unit to start supplying the fuel gas;
The fuel cell system further includes an impurity amount parameter measurement unit that measures an impurity amount parameter value correlated with the amount of impurities in the plurality of internal fuel gas flow paths,
The control unit measures the impurity amount parameter value using the impurity amount parameter measurement unit at the start-up,
(I) When the impurity amount parameter value indicates that the impurity amount is relatively large, the fuel gas is supplied according to the first supply mode;
(Ii) When the impurity amount parameter value indicates that the impurity amount is relatively small, the fuel gas is supplied according to the second supply mode.
Fuel cell system. The fuel cell system according to claim 1 or 2, further comprising:
Exhaust gas discharge flow paths from the plurality of internal fuel gas flow paths to the outside from the plurality of internal fuel gas flow paths,
An openable / closable discharge valve provided in the discharge channel;
With
The control unit executes a discharge process of discharging the exhaust gas to the outside by opening the discharge valve after starting the fuel gas supply according to the first supply mode at the time of starting the fuel cell system. ,
Fuel cell system. The fuel cell system according to claim 3, wherein
The control unit causes the fuel gas supply unit to stop the supply of the fuel gas before the discharge process, and executes the discharge process in a state where the supply of the fuel gas is stopped. The fuel cell system according to claim 3 or 4, further comprising:
A differential pressure parameter measurement unit for measuring a differential pressure parameter value correlated with a differential pressure between the pressure on the upstream side and the pressure on the downstream side of the discharge valve;
In the discharging process, the control unit measures the differential pressure parameter value using the differential pressure parameter measurement unit, and the differential pressure parameter value indicates that the differential pressure is equal to or less than a predetermined allowable differential pressure. A fuel cell system in which the exhaust valve is kept open until the indicated value is reached. A method for starting a fuel cell system including a fuel cell having a plurality of internal fuel gas flow paths,
(A) a step of supplying the fuel gas to the plurality of internal fuel gas flow paths in a first supply mode as a supply mode for supplying fuel gas when the fuel cell system is activated;
The step (A)
(A1) starting a process of circulating the exhaust gas from the plurality of internal fuel gas passages again to the plurality of internal fuel gas passages;
(A2) starting supply of the fuel gas to the plurality of internal fuel gas flow paths at a timing when a predetermined circulation time greater than zero has elapsed since the start of circulation;
A starting method.

Images