JP2007027078A - Fuel cell system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a fuel cell system for preventing degradation of a cathode electrode at start-up. <P>SOLUTION: The fuel cell system equipped with a bypass flow channel 6 circulating exhaust anode gas to an anode is provided with a gas composition control means 7 controlling a gas composition of mixture gas consisting of hydrogen and the exhaust anode gas supplied to the anode 1a, a gas exhaust control means 8 controlling exhaust outside of the exhaust anode gas, and a controller 20 controlling the gas composition control means 7 and the gas exhaust control means 8. The controller 20 closes a gas exhaust control valve 8 at the start-up when oxidant gas exists inside the anode 1a to circulate the exhaust anode gas to the anode 1a, and controls the gas composition control means 7 so that a density of gas constituting the mixture gas is at a given level. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a fuel cell system.

  An ordinary polymer electrolyte fuel cell is composed of an electrolyte membrane / electrode assembly (hereinafter referred to as a polymer ion exchange membrane) having an anode and a cathode sandwiched between both sides of a polymer ion exchange membrane (cation exchange membrane). And MEA) are further sandwiched between separators. The fuel gas (hydrogen) supplied to the anode is hydrogen ionized on the catalyst electrode and moves to the cathode through the electrolyte membrane that is appropriately humidified. Electrons generated in the meantime are taken out to an external circuit and used as direct current electric energy. Since an oxidant gas such as an oxygen-containing gas or air is supplied to the cathode, the hydrogen ions, the electrons, and the oxygen gas react with each other to generate water at the cathode. The chemical formula showing the above reaction is as follows.

Anode reaction: H ₂ → 2H ⁺ + 2e ⁻ (1)
Cathode reaction: 2H ⁺ + 2e ⁻ + (1/2) O ₂ → H ₂ O (2)
One set of the polymer electrolyte fuel cells is called a unit cell. By stacking a predetermined number of these unit cells, a stacked fuel cell is configured. The catalyst electrode is composed of carbon (carbon) as a carrier and, for example, a platinum catalyst as a reaction catalyst supported on the carbon.

  However, when the fuel cell 1 is in a no-load state (the current is not flowing through the circuit connected to the fuel cell 1; the same applies hereinafter), when the battery cell 1 is left or started, the carbon of the cathode is deprived and the MEA deteriorates. Will do. This carbon corrosion will be described with reference to FIG.

In a conventional fuel cell system, when the fuel cell is stopped in a no-load state or when hydrogen supply to the anode is started from the start-up (when hydrogen and oxygen are separated in the anode), the anode to the cathode The proton H ⁺ moves to the water, and the transferred proton H ⁺ reacts with oxygen at the cathode to produce water (see the above formula). Proton H ⁺ is required for this reaction, and in the no-load state, electrons e ⁻ do not move through the load. For this reason, water present in the cathode reacts with carbon as a carrier, and the generated electrons e ⁻ are used in the water generation reaction described above. During this reaction, the carbon of the cathode is consumed and the MEA deteriorates. On the other hand, at the anode, oxygen mixed with hydrogen, proton H ⁺ generated when carbon is consumed, and electron e ⁻ generated by protonation of hydrogen react to generate water.

Here, when the open-circuit voltage is high, the movement of electrons e ⁻ becomes active, the above-described chemical reaction is promoted, and the MEA is further deteriorated. As the MEA deteriorates, the power generation voltage decreases and the power generation efficiency decreases.

In order to prevent such deterioration of the MEA, a fuel gas (hydrogen) having a high pressure is flowed into the anode, and oxygen present in the anode is discharged to the outside, thereby mixing the fuel gas and oxygen in the fuel cell. There is a technique for suppressing deterioration of MEA by limiting the state to a short time (see Patent Document 1).
JP 2004-139984 A

  However, in the technique described in Patent Document 1, when oxygen is discharged to the outside, hydrogen is also discharged together, and there is a problem that hydrogen is discharged wastefully and the power generation efficiency of the fuel cell system is reduced. .

  Accordingly, an object of the present invention is to prevent the deterioration of the MEA at the time of starting the fuel cell, prevent the release of hydrogen, and suppress the decrease in the efficiency of the fuel cell.

  The present invention provides a hydrogen supply channel for supplying hydrogen from the outside to the anode of the fuel cell, and a bypass channel for connecting exhaust hydrogen gas discharged from the anode to the anode, connected to the hydrogen supply channel. And a gas composition control means for controlling a gas composition of a mixed gas composed of the hydrogen and exhaust anode gas supplied to the anode, and a gas exhaust control for controlling discharge of the exhaust anode gas to the outside. And a controller for controlling the gas composition control means and the gas discharge control means, and at the start-up when the oxidant gas is present in the anode, the controller closes the gas discharge control valve, and Is circulated to the anode through the bypass flow path so that the concentration of the gas constituting the mixed gas becomes a predetermined concentration. To control the control means.

  In the present invention, when the fuel cell system is started, the discharge of the exhaust anode gas is prohibited, the exhaust anode gas is circulated to the anode, and the composition of the mixed gas of the exhaust anode gas and external hydrogen is controlled to a predetermined concentration. Therefore, hydrogen is not released to the outside during startup, and corrosion of the cathode can be prevented.

  FIG. 1 is a configuration diagram of an embodiment of a fuel cell system to which the present invention is applied. This fuel cell system includes a stacked fuel cell 1, a hydrogen tank 2 that supplies hydrogen to the anode 1a of the fuel cell, a compressor 10 that supplies air to the cathode 1b, and exhausted anode gas discharged from the anode 1a. Is provided with a bypass flow path 6 that circulates to the anode 1a, and a circulation pump 7 that is installed in the bypass flow path 6 and controls the circulation amount of the exhaust anode gas.

  Hydrogen is supplied to the anode 1a of the fuel cell 1 from a hydrogen tank 2 storing high-pressure hydrogen through a hydrogen supply channel 2a, and air is supplied to the cathode 1b through the air supply channel 3a by the action of the compressor 10. Hydrogen or air used for power generation in the fuel cell 1 is discharged to the outside from the gas discharge channel 2b and the air discharge channel 3b. A purge valve 8 for controlling hydrogen released to the outside from the gas discharge channel 2b is installed in the gas discharge channel 2b.

  The hydrogen supply passage 2a is supplied to a hydrogen tank main valve 3 that controls the supply of hydrogen from the hydrogen tank 2, a pressure reducing valve 4 that reduces the pressure of hydrogen from the hydrogen tank 2 to a predetermined pressure, and an anode 1a. A hydrogen supply control valve 5 for controlling the amount of hydrogen is installed. Further, a bypass passage 6 is formed which connects the hydrogen supply control valve 5 downstream side (fuel cell 1 side) of the hydrogen supply passage 2 a and the purge valve 8 upstream of the gas discharge passage 2 b by bypassing the fuel cell 1. Is done. The bypass passage 6 is provided with a circulation pump 7 for circulating the exhaust anode gas discharged from the fuel cell 1 to the anode 1a.

  Therefore, when the circulation pump 7 of the bypass flow path 6 is operated with the purge valve 8 installed in the gas discharge flow path 2b closed, the exhaust anode gas discharged from the anode 1a of the fuel cell 1 is 6 is supplied again to the anode 1a. On the other hand, an air pressure regulating valve 9 for adjusting the air pressure of the cathode 1b is installed in the air discharge passage 3b.

  When hydrogen is supplied to the anode 1a and air is supplied to the cathode 1b, power generation is started by the fuel cell 1. The power generated by the fuel cell 1 is taken out by the power manager 13 and used as a drive source for a load, for example, a moving body. Supplied to the motor.

  A controller 20 for integrated control of the fuel cell system is installed. The controller 20 includes an output of a first pressure sensor 11 a that detects the pressure of the mixed gas of hydrogen and exhaust anode gas at the anode inlet, and a second pressure sensor 11 b that detects the pressure of hydrogen flowing into the hydrogen supply control valve 5. The output of the hydrogen concentration sensor 12 for detecting the hydrogen concentration at the anode inlet and the output of the voltage sensor 14 for detecting the power generation amount of the fuel cell 1 are input.

  The controller 20 uses these input values to control the start-up of the fuel cell 1, the opening / closing control of the hydrogen tank main valve 3, the flow control of the hydrogen supply control valve 5, the operation control of the compressor 10, the operation control of the circulation pump 7, and the purge It controls the opening and closing control of the valve 8, and the pressure control of the air pressure control valve 9.

  The controller 20 further receives an on / off signal of the main switch 15 that activates the fuel cell system and an on / off signal of the load switch 19 that detects the load state of the fuel cell 1.

  FIG. 2 is a block diagram illustrating the configuration of the controller 20. The controller 20 includes an activation request determination unit 21 that determines an activation request of the fuel cell 1 based on an on / off signal of the main switch, and detection values of the first and second pressure sensors 11a and 11b, the concentration sensor 12, and the voltage sensor 14 described above. The operation state detection unit 22 that detects the operation state of the fuel cell 1 from the input, the start request determination result of the fuel cell 1 and the operation state of the fuel cell 1 are input, and based on these inputs, the start time of the fuel cell 1 is Similarly, a startup calculation unit 23 that calculates operating conditions, a fuel agent supply means control unit 24 that controls the hydrogen supply control valve 5 based on the calculation result of the startup calculation unit 23 and controls the hydrogen supply amount to the anode 1a, Based on the calculation result of the start calculation unit 23, it is constituted by a fuel agent circulation means control unit 25 that controls the circulation pump 7 and controls the amount of exhaust anode gas that circulates in the bypass flow path 6.

  FIG. 3 is a flowchart for explaining the control contents performed by the controller 20 when starting the fuel cell system. This start-up control is control that is performed at the time of start-up from a no-load state in which an oxidant gas (air) is present in the anode 1 a and no current is flowing through the circuit connected to the fuel cell 1.

  The oxidant gas present in the anode 1a when the fuel cell 1 is started is air or the like that has entered from the outside during the stop of the fuel cell. Therefore, the startup control is performed on the assumption that air is present in the anode 1a during startup.

  First, in step S11, it is determined from the on / off signal of the main switch 15 whether or not there is a request for starting the fuel cell system. If it is on, it is determined that there is an activation request and the process proceeds to step S12. Note that at the time of startup, the fuel cell 1 is controlled so as to maintain a no-load state. In step S12, the purge valve 8 that shields the gas discharge passage 2b connected to the anode 1a of the fuel cell 1 is closed. In subsequent step S13, the hydrogen tank main valve 3 is opened, and in step S14, the circulation pump 7 is operated. As a result, the exhaust anode gas circulates through the bypass flow path 6 to the anode 1a.

  The circulation pump in step S14 prevents the hydrogen supplied from the hydrogen tank 2 to the anode 1a through the opening of the hydrogen supply control valve 5 in the next step S15 from flowing into the bypass channel 6 from the hydrogen supply channel 2a. Circulate the exhaust anode gas.

  In step S15, the hydrogen supply control valve 5 is opened, and hydrogen is supplied from the hydrogen tank 2 to the anode 1a. Accordingly, by supplying hydrogen from the hydrogen tank 2 to the anode 1a after the circulation pump 7 for circulating the exhaust anode gas is operated, the oxidant gas in the anode 1a is mixed with hydrogen, and the bypass channel from the gas discharge channel 2b. 6 is sent.

  In the subsequent step S16, the circulation amount of exhaust anode gas to the anode 1a is controlled by the circulation pump 7. The amount of exhaust anode gas circulated to the anode 1a is set according to the hydrogen concentration in the mixed gas detected by the concentration sensor 12 installed at the anode inlet. Specifically, the target ratio of the concentration of hydrogen from the hydrogen tank 2 and the bypass passage 6 to the concentration of air from the bypass passage 6 supplied to the anode 1a is high in hydrogen, for example, hydrogen 7: The operation load of the circulation pump 7 is feedback-controlled so that the oxidant gas (air) 3 is obtained.

  In subsequent step S17, it is determined whether or not the elapsed time from the start of hydrogen supply in step S15 has reached a predetermined time. If the predetermined time has elapsed, the process proceeds to step S18. If not, the process returns to step S16 to repeat the circulation pump control. Here, the predetermined time is a time taken until all the oxygen in the anode 1a is consumed by the reaction with hydrogen. The amount of hydrogen supplied to the anode 1a, the concentration of hydrogen supplied to the anode 1a, and the air Estimate the amount of oxygen from the target ratio to the concentration (for example, 7: 3), integrate the estimated amount of oxygen, and until the integrated amount of oxygen reaches the maximum amount that can exist in the volume of the hydrogen supply channel 2a. Let time be a predetermined time.

  Then, in step S18, it is determined that all the oxygen in the anode 1a has been consumed because a predetermined time has elapsed, the fuel cell 1 is started, the purge valve 8 is opened in step S19, and the routine proceeds to normal operation. The current flows between the load and the fuel cell 1 only after the fuel cell 1 is started. Until step S18 when the fuel cell 1 is activated, the fuel cell system operates with a secondary battery (not shown) as a power source.

  Next, a method for setting the target ratio between the hydrogen concentration and the air concentration in the mixed gas supplied to the anode 1a will be described. The inventors paid attention to the relationship between the concentration ratio of hydrogen and air (oxygen) in the mixed gas supplied to the anode 1a and the carbon corrosion of the cathode 1b, and flowed mixed gases having different concentration ratios to the anode 1a. In this case, the occurrence of carbon corrosion of the cathode 1b was confirmed by experiments.

  FIG. 4 is a diagram showing a state of occurrence of carbon corrosion at the cathode 1b when a mixed gas having various concentration ratios of hydrogen and air is introduced into the anode 1a. Here, the carbon corrosion can be determined based on the amount of carbon dioxide generated during the carbon corrosion.

  First, carbon dioxide generated when a gas having a hydrogen concentration of 100% is introduced into the anode 1a is temporarily generated when the boundary between the supplied hydrogen and the air in the anode 1a passes through the anode 1a. . Further, when a mixed gas having a hydrogen concentration of 70% + oxygen concentration of 30% was introduced into the anode 1a, no significant change was observed in the amount of carbon dioxide generated when a hydrogen concentration of 100% was introduced.

  However, when a mixed gas having a hydrogen concentration of 40% + oxygen concentration of 60% was introduced into the anode 1a, the amount of carbon dioxide generated was increased about four times as compared with the case where a hydrogen concentration of 100% was introduced. Further, it has been found that carbon dioxide is continuously generated when a mixed gas having a hydrogen concentration of 30% and an oxygen concentration of 70% is introduced into the anode 1a. In addition, the point in a figure plots the generation amount for 3 minutes after a hydrogen supply start. From this result, it can be seen that the carbon corrosion of the cathode 1b can be suppressed by setting the composition of the mixed gas supplied to the anode 1a to a hydrogen concentration of 70% or more.

  FIG. 5 shows the results of examining the hydrogen concentration and the air concentration of the exhaust anode gas discharged from the anode 1a with respect to the hydrogen concentration of the mixed gas of hydrogen and air introduced into the anode 1a. According to this, when carbon dioxide is continuously generated in FIG. 5 (in the case of a mixed gas having a hydrogen concentration of 30% + oxygen concentration of 70%), it can be seen that hydrogen is not contained in the exhaust anode gas. That is, by controlling the exhaust anode gas to contain hydrogen, it is possible to suppress the continuous generation of carbon dioxide, that is, the progression of continuous carbon corrosion. On the other hand, by setting the hydrogen concentration of the mixed gas to 30% or more, the oxygen concentration in the exhaust anode gas becomes almost 0%, and oxygen in the anode 1a can be removed.

  Therefore, in the present embodiment, when the fuel cell system is activated, the purge valve 8 is set when the oxidant gas is present in the anode 1a and the current connected to the circuit connected to the fuel cell 1 is not loaded. Then, the circulation pump 7 is started and the hydrogen supply control valve 5 is opened in a state in which the exhaust anode gas discharged from the anode 1a is circulated through the bypass flow path 6 to supply hydrogen from the hydrogen tank 2 to the anode 1a. Then, the circulation pump 7 is controlled so that the hydrogen concentration of the mixed gas of the hydrogen supplied to the anode 1a and the exhaust anode gas becomes a predetermined value (for example, 70%) or more.

  In this way, the composition (concentration) of the gas constituting the mixed gas supplied to the anode 1a is controlled to suppress the carbon corrosion of the cathode 1b, and the exhaust anode gas containing hydrogen discharged from the anode 1a at the start-up. Is circulated to the anode 1a through the bypass flow path 6 and useless release of hydrogen is prohibited, so that a decrease in power generation efficiency of the fuel cell system can be suppressed. In addition, the hydrogen supply control valve 5 is opened after the purge valve 8 is closed and the circulation pump 7 is started at the time of activation, so that the hydrogen from the hydrogen tank 2 is prevented from flowing back into the bypass passage 6. Note that the hydrogen supply control valve 5 may be used together with the circulation pump 7 to control the hydrogen concentration of the mixed gas supplied to the anode 1a.

  FIG. 6 is a flowchart for explaining the control contents of the controller 20 as the second embodiment. The difference from the first embodiment is that the oxygen concentration in the mixed gas supplied to the anode 1a is changed to the hydrogen concentration sensor 12 for detecting the hydrogen concentration in the mixed gas supplied to the anode 1a. The oxygen concentration sensor to detect is provided, and the other configuration is the same as that of the first embodiment. Hereinafter, the difference in the control contents in this embodiment will be described in comparison with the flowchart of the first embodiment.

  Steps S11 to S16 are the same control contents as in the first embodiment. In step S27 following step S16, the oxygen concentration is read from the newly set oxygen concentration sensor, and it is determined whether or not the detected oxygen concentration is less than a predetermined concentration. Here, for example, when the oxidant gas remaining in the anode 1a is air, the predetermined concentration is set to 6%. When the detected oxygen concentration is 6% or more, the predetermined concentration is circulated so as to be less than 6%. The operation load of the pump 7 is controlled in step S16.

  If the oxygen concentration is less than the predetermined concentration, the process proceeds to step S28 and subsequent steps, the fuel cell 1 is started in step S28, and the power generation amount of the fuel cell 1 according to the user's request is circulated in step S29. The operation load of the pump 8 and the opening degree of the hydrogen supply control valve 5 are controlled. In step S30, the purge valve 8 is opened to release the residual nitrogen in the anode 1a to the outside.

  In this embodiment, the time until the purge control in step S30 can be shortened while providing the same effects as in the first embodiment.

  In addition, by providing the hydrogen concentration sensor 12 of the first embodiment and the oxygen concentration sensor of the second embodiment together, the ratio between the hydrogen concentration and the oxygen concentration is further increased in consideration of the amount of nitrogen and water vapor in the air. It can be controlled with high accuracy.

  In the first and second embodiments, the concentration sensor is installed at the anode inlet, but a concentration sensor for detecting the carbon dioxide concentration may be installed at the cathode outlet. In this case, as shown in FIG. 12, carbon dioxide is generated due to the deterioration of the carbon. Therefore, the deterioration of the carbon is determined from the concentration of the generated carbon dioxide so that the carbon dioxide becomes a predetermined concentration or less. Further, the operation load of the circulation pump 8 is controlled to prevent the MEA from being deteriorated.

  A concentration sensor that detects the oxygen concentration may be installed at the anode outlet. This is because when the circulation pump 7 is operated when air does not enter the fuel cell from the outside, such as immediately after the fuel cell 1 stops, but the air enters the bypass channel 6, the anode 1 a A mixed gas having a high oxygen concentration is supplied to the anode 1a. When such a gas is supplied to the anode 1a, oxygen that is not used in the reaction is generated in the anode 1a, and the excess oxygen is discharged to the gas discharge channel 2b. The concentration of the discharged oxygen is detected by a sensor, and the operation load of the circulation pump 7 is set so that the detected oxygen concentration is lower than a predetermined concentration.

  Furthermore, the operation load of the circulation pump 7 can be controlled using the output value of the voltage sensor 14 instead of the hydrogen concentration sensor 12 of the first embodiment. When an oxygen-rich mixed gas having a high oxygen concentration is supplied to the anode 1a, the power generation voltage of the fuel cell 1 decreases. Therefore, when the voltage of the single cell constituting the fuel cell 1 or the voltage for each predetermined number of single cells is detected using the voltage sensor 14 and the change amount of the voltage is equal to or greater than the predetermined change amount, the circulation pump 7 Control is performed to supply a mixed gas with a high hydrogen concentration by reducing the discharge amount.

  FIG. 7 is a configuration diagram of the third embodiment. The characteristic configuration of this embodiment includes an oxygen concentration sensor 16 that detects the oxygen concentration upstream of the circulation pump 7 in the bypass flow path 6 (on the gas release flow path 2b side), and a downstream of the circulation pump 7 (hydrogen supply flow path). 2a), a circulation amount control valve 18 for controlling the circulation amount of the exhaust anode gas in the bypass passage, and a bypass passage between the circulation pump 7 and the circulation amount control valve 18 are provided. And a third pressure sensor 11c for detecting the internal pressure. The outputs of the sensors 11c and 16 are output to the controller 20, and the circulation amount control valve 18 is controlled by the controller 20 based on the outputs of the sensors 11c and 16.

  FIG. 8 is a flowchart for explaining the control contents of the present embodiment performed by the controller 20. Steps S11 to S13 are the same as the control in the first embodiment, and a description thereof will be omitted.

  After the hydrogen tank main valve 3 is opened in step S13, the circulation amount control valve 18 installed in the bypass flow path 6 is closed in the subsequent step S31. By closing the circulation amount control valve 18, hydrogen is prevented from flowing into the bypass flow path 6 from the hydrogen supply flow path 2a. In the subsequent step S32, the circulation pump 7 is operated. Thereby, the pressure of the bypass flow path 6 between the circulation amount control valve 18 and the circulation pump 7 increases. At this time, the pressure of the bypass flow path 6 detected by the third pressure sensor 11c is higher than the pressure of the hydrogen supply flow path 2a detected by the first pressure sensor 11a.

  In a succeeding step S33, it is determined whether or not the detected concentration is lower than a predetermined concentration using the detection value of the oxygen concentration sensor 16. Here, the predetermined concentration is, for example, a value of 6% or less that can be determined that the oxygen concentration in the exhaust anode gas has become sufficiently low. If it is less than the predetermined value, it is determined that the oxygen concentration in the exhaust anode gas is sufficiently low, and the process proceeds to step S37. If it is equal to or greater than the predetermined value, the process proceeds to step S34.

  In step S34, the hydrogen supply control valve 5 is opened by a predetermined amount, and supply of hydrogen from the hydrogen tank 2 to the anode 1a is started. In the subsequent step S35, the oxygen concentration in the exhaust anode gas is detected again using the oxygen concentration sensor 16, and the detected oxygen concentration is compared with the predetermined concentration used in step S33 to determine whether it is less than the predetermined concentration. If the detected concentration is less than the predetermined concentration, the process proceeds to step S37. If the detected concentration is equal to or greater than the predetermined concentration, the process proceeds to step S36. In step S36, using the circulation amount control valve 18, the exhaust anode gas is adjusted so that the oxygen concentration is less than the predetermined concentration. Control the amount of circulation. At this time, the circulation pump 7 is controlled so that the pressure upstream of the circulation amount control valve 18 increased in step S32 does not decrease.

  In step S37, assuming that the oxygen concentration in the anode 1a has become sufficiently low, the circulation amount control valve 18 is fully opened, and the routine proceeds to step S28 to shift to normal operation. Steps S28 to S30 are the same as those in the first embodiment.

  In the case of this embodiment, since the circulation amount of the exhaust anode gas flowing through the bypass flow path 6 is controlled by the circulation amount control valve 18 so that the oxygen concentration in the anode 1a is sufficiently low, hydrogen supply to the anode 1a is performed. Even when the change in the amount is fast, the circulation amount can be changed rapidly in response to the change in the hydrogen supply amount.

  In addition, when hydrogen is supplied at the time of starting the fuel cell system, initially, the pressure in the anode 1a is low and the difference from the original pressure of the hydrogen tank 2 (or the pressure downstream of the pressure reducing valve 4) is large. In many cases, if the hydrogen supply is continued, the pressure in the anode 1a increases, and the hydrogen supply amount decreases. On the other hand, if the responsiveness of the circulation pump 7 is poor, the supply of air will not be in time at the beginning of the hydrogen supply. Therefore, in the third embodiment, before supplying hydrogen in step S34, the circulation amount control valve 18 is closed and the circulation pump 7 is operated to bypass the circulation flow path between the circulation pump 7 and the circulation amount control valve 18. By storing high-pressure air in 6, a large amount of air can be supplied immediately after supplying hydrogen, so that the air can be reacted quickly and consumed.

  In the third embodiment, the oxygen concentration sensor 16 is used to measure the air (oxygen) concentration in the hydrogen supply path immediately after startup (step S35). The air concentration may be estimated by measuring the time until the activation request is received. Since the amount of air entering the hydrogen supply channel 2a enters by natural diffusion and depends on time, it is possible to estimate the amount of air. Here, the stop of the fuel cell means a state in which the power extraction is stopped and the purge valve is opened.

  Further, the air concentration may be estimated from the power consumption of the circulation pump 7. The rotation speed of the circulation pump immediately after start-up is always kept constant at start-up, the relationship between the rotation speed and power consumption at that time is detected, and if the air concentration in the exhaust gas to be circulated is high, the power consumption is large and hydrogen If the concentration is high, the power consumption is small. Therefore, this tendency can be used to estimate the air concentration.

  FIG. 9 is a configuration diagram of the fourth embodiment. Compared to the configuration of the first embodiment, a differential pressure sensor 17 that detects a pressure difference between the upstream and downstream of the circulation pump 7, and a differential pressure sensor 17. The fourth pressure sensor 11d is provided upstream of the first pressure sensor. The output signals of these sensors 17 and 11d are sent to the controller 20.

  FIG. 10 is a flowchart for explaining the control contents of the present embodiment.

  The contents from step S11 to step S15 are the same as the control contents of the first embodiment, and a description thereof will be omitted.

  In the subsequent step S41, the amount of hydrogen supplied to the anode 1a is calculated. The hydrogen supply amount is calculated based on the detection of the first and second pressure sensors 11 a and 11 b that detect the pressures on the upstream side and the downstream side of the hydrogen supply control valve 5. Further, the flow rate of the exhaust anode gas circulating from the bypass flow path 6 to the anode 1a is calculated based on the calculated hydrogen supply amount. The circulation flow rate of the exhaust anode gas is controlled by the discharge amount of the circulation pump 7 so as to be 30% or less with respect to the calculated hydrogen supply amount. The target discharge amount of the circulation pump 7 is determined based on the detected values of the installed differential pressure sensor 17 and the fourth pressure sensor 11d and the performance of the circulation pump 7. Here, the flow rate indicates a mass flow rate. In the present embodiment, control is performed using various sensors, but the pressure loss of the flow path through which the fuel cell 1 and hydrogen flow is estimated without using the sensor, and the target discharge amount of the circulation pump 7 is set. Also good.

  As shown in FIG. 11, the hydrogen supply amount tends to decrease as the anode inlet pressure increases, and the discharge amount of the circulation pump 7 is controlled in accordance with this tendency.

  In this embodiment, the supplied hydrogen supply amount is calculated from the pressure difference before and after the hydrogen supply control valve 5, and the exhaust anode circulated from the bypass flow path 6 to the anode 1a according to the calculated hydrogen supply amount. By setting the gas flow rate, it is possible to prevent the corrosion of carbon by managing the hydrogen flow rate and the exhaust anode gas flow rate at the anode inlet without detecting the gas concentration such as hydrogen or air.

  In the embodiments so far, the circulation pump 7 is installed to circulate the exhaust anode gas through the bypass flow path 6. However, instead of the flow rate variable control of the circulation pump 7, an ejector that supplies hydrogen into the bypass flow path 6. May be provided.

  Therefore, in the present embodiment, in the fuel cell system having a bypass flow path for circulating the exhaust anode gas exhausted from the anode 1a supplied with hydrogen from the outside to the anode 1a, the hydrogen supplied to the anode 1a and the exhaust gas are exhausted. A gas composition control means for controlling the gas composition of the mixed gas comprising the anode gas; a gas discharge control means for controlling the discharge of the exhaust anode gas to the outside; and the gas composition control means and the gas discharge control means are controlled. Provided with a controller, when the oxidant gas is present in the anode 1a, and when starting from a no-load state in which no current flows in a circuit connected to the fuel cell, the controller closes the gas discharge control valve, The gas composition control means is controlled so that the concentration of the gas constituting the mixed gas supplied to the anode 1a becomes a predetermined concentration. To order, the exhaust anode gas by circulating through the bypass passage when starting, the hydrogen can reduce a decrease in the power generation efficiency of the fuel cell system to prevent from being emitted to the outside. Further, since the gas composition control means is controlled so that the hydrogen concentration of the mixed gas supplied to the anode 1a is equal to or higher than the predetermined concentration, carbon corrosion at the cathode 1b can be prevented and deterioration of the fuel cell can be suppressed.

  FIG. 13 is a configuration diagram of the fifth embodiment.

  Compared with the configuration of the first embodiment shown in FIG. 1, the present embodiment is that the hydrogen concentration sensor 12 is changed to the oxygen concentration sensor 31, and the gas discharge flow path 2 b is released from the purge valve 8 at the downstream side. The difference is that the upstream side of the air pressure regulating valve 9 in the flow path 3b is joined. Further, the hydrogen tank main valve 3, the pressure reducing valve 4, and the first pressure sensor 11a are eliminated from the configuration.

  With this configuration, the present embodiment is characterized in that the exhaust anode gas circulated through the bypass passage 6 is circulated to the anode with the maximum circulation amount that can be stably controlled.

  FIG. 14 is a flowchart for explaining the contents of control performed by the controller 20 when the fuel cell system according to this embodiment is started. This start-up control is a control that is performed during start-up from a no-load state in which oxidant gas (air) is present in the anode 1a, the hydrogen supply flow path 2a, and the bypass flow path 6.

  The oxidant gas present in the anode 1a when the fuel cell 1 is started is air that has entered from the outside while the fuel cell is stopped, air that has cross-leaked from the cathode 1b to the anode 1a through the electrolyte membrane, or the like. is there.

  Further, the state before the start-up control of the present embodiment is started (that is, when the fuel cell system is stopped) is a state in which all the components are stopped. The purge valve 8 is opened, the hydrogen supply control valve 5 is closed, and the air pressure regulating valve 9 is opened.

  First, in step S51, it is determined whether or not a trigger for determining activation of the fuel cell system is on. As a trigger for the activation determination, for example, it is assumed that the main switch 15 is in an on state. If the trigger is on, the process proceeds to step S52, and if it is off, the control is terminated. In step S52, the purge valve 8 is closed so that the exhausted anode gas discharged from the anode is circulated to the anode through the bypass passage 6. By closing the purge valve 8, hydrogen supplied in step S57 described later is not released to the outside. In step S53, the compressor 10 is activated. The compressor 10 is controlled to a rotational speed corresponding to an idle state described later.

  Next, the circulation pump 7 is started at step S54. Here, the rotation speed of the circulation pump 7 is controlled to the maximum rotation speed during normal operation that can be stably controlled. In step S55, the rotational speed of the circulation pump 7 is detected by a rotational speed sensor or the like (not shown), and it is determined whether or not it has converged stably to the maximum rotational speed. If it has converged, the process proceeds to step S56. If it has not converged, the control is maintained until it converges, and the determination is repeated. In step S56, the opening degree of the air pressure regulating valve 9 is reduced, and pressure increase in the cathode is started. Subsequently, in step S57, the hydrogen supply control valve 5 is opened, and supply of hydrogen to the anode of the fuel cell 1 is started.

  Next, in step S58, it is determined whether or not a predetermined time has elapsed since the start of hydrogen supply. Here, the predetermined time is set to the predetermined time obtained by obtaining in advance an experiment or the like the time when hydrogen is expected to reach the anode. When the predetermined time has elapsed, the process proceeds to step S59, where current extraction from the fuel cell 1 is started. Here, a minute load current corresponding to the idle state is taken out.

  Subsequently, in S60, the output of the oxygen concentration sensor 31 is read to determine whether or not the detected oxygen concentration has reached a predetermined value (for example, 6%) or less, and when it becomes 6% or less, the process proceeds to S61. Here, the concentration is set to 6%. However, according to a deterioration experiment result described later, the concentration may be 6% or less. In S61, the rotational speed of the circulation pump 7 is reduced to a rotational speed corresponding to the idle state, the start-up control is terminated, and the process proceeds to a normal operation. Here, the number of revolutions corresponding to the idle state is the number of revolutions in the idle state in which the fuel cell system is in a power generation enabled state and is not in a power supply state for supplying power to an external load or the like.

  In the present embodiment, the expression of the maximum number of rotations of the circulation pump 7 is used. However, it is not limited to one number of rotations, and it goes without saying that the number of rotations near the maximum number of rotations is included. For example, in FIG. 15, the actual maximum rotational speed of the circulation pump 7 is 12000 (rpm), but the maximum rotational speed for explanation of the present embodiment is defined within a range of, for example, 10000 to 12000 (rpm).

  FIG. 15 shows the relationship between the carbon dioxide emission measured at the cathode outlet and the rotational speed of the circulation pump 7. This is the case where the hydrogen supply control valve 5 is opened to start hydrogen pressure increase and hydrogen is supplied to the anode of the fuel cell 1 in a state where the rotation speed of the circulation pump 7 is stably controlled to a predetermined rotation speed. The relationship between the discharge amount of carbon dioxide from the cathode and the rotational speed of the circulation pump 7 is shown.

As a result, a constant flow rate of air is allowed to flow through the cathode of the fuel cell 1, and the amount of carbon dioxide contained in the air discharged from the cathode by the gas analyzer provided at the cathode outlet (included when air is supplied). (Excluding CO ₂ ).

  When only the rotation speed of the circulation pump 7 is changed to 4000, 8000, and 12000 (rpm) and the other test conditions are the same, the amount of carbon dioxide discharged from the cathode outlet is remarkably increased when the rotation speed of the circulation pump 7 is increased. It turns out that it reduces. This emission of carbon dioxide is considered to be caused by carbon corrosion at the cathode shown in the following formula. That is, by increasing the number of revolutions of the circulation pump 7, the circulation amount of hydrogen containing hydrogen (exhaust anode gas) is increased, and the remaining air is quickly consumed at the anode, and the carbon corrosion expressed by the following equation is generated at the cathode. Since the reaction is suppressed, carbon dioxide emissions are reduced.

C + 2H ₂ O → CO ₂ + 4H ⁺ + 4e ⁻ (3)
Next, air discharged from the cathode when a constant flow rate of air is supplied to the cathode of the single cell constituting the fuel cell 1 and a mixed gas of hydrogen and air mixed at a predetermined ratio is supplied to the anode ( The results of measuring the amount of carbon dioxide in the exhaust cathode gas) are shown in FIG.

  According to this result, the amount of carbon dioxide exhausted from the cathode varies greatly depending on the hydrogen / air concentration ratio supplied to the anode, and hydrogen / air = 10/90% to 70/30%. Emission is confirmed, and it can be seen that the emission of carbon dioxide becomes almost zero when the hydrogen concentration is 70% or more. Here, since hydrogen / air = 70/30% corresponds to hydrogen / nitrogen / oxygen = 70/24/6%, the predetermined value 6% of the oxygen concentration in step S60 is set. This carbon dioxide generation is also the same reaction mechanism as the above equation (3), and is an indicator of cathode deterioration. In this embodiment, the carbon dioxide emission determination is performed using the oxygen concentration. However, the determination may be performed using the hydrogen concentration instead of the oxygen concentration.

  In the stage before hydrogen supply at startup, oxygen may be contained in the anode of the fuel cell 1, the hydrogen supply flow path 2a as the hydrogen supply path, and the bypass flow path 6. When hydrogen is supplied from such a state, a mixed gas in which hydrogen and oxygen coexist flows in the anode and the hydrogen supply paths 2a and 6. This mixed gas is circulated to the anode as the exhaust anode gas through the bypass channel 6, and there is a risk that oxidation deterioration of the cathode (corrosion of the carbon support, elution of catalyst metal, grain growth, etc.) may proceed. Therefore, in this embodiment, by supplying hydrogen to the anode with the maximum circulation amount that can be realized in the fuel cell system, the remaining oxygen is consumed quickly in a short time due to the combustion reaction at the anode. The oxidation deterioration of the cathode as shown in the formula can be minimized.

  In addition, by setting the rotation speed of the circulation pump 7 in the vicinity of the maximum rotation speed within the range that can be stably controlled, the residual oxygen is consumed quickly in a short time due to the combustion reaction at the anode. Oxidation deterioration of the cathode can be suppressed by the control.

  In addition, after the introduction of hydrogen, when the oxygen concentration in the vicinity of the anode and in the hydrogen supply path falls below a predetermined concentration close to 0, the oxidation deterioration reaction of the cathode is suppressed, so the circulation rate is reduced from the viewpoint of cathode deterioration. There is no problem even if you let it. Therefore, the power of the circulation pump 7 can be reduced and power consumption can be suppressed.

  FIG. 17 is a configuration diagram of the sixth embodiment.

  This embodiment is different from the configuration of the fifth embodiment shown in FIG. 13 in that the oxygen concentration sensor 31 is deleted. Instead of detecting the oxygen concentration, the time until the oxygen concentration reaches the predetermined oxygen concentration from the start of hydrogen supply is obtained in advance by experiments, etc., and when it reaches the predetermined time, the start-up control is terminated and the operation proceeds to normal operation. It is what I did.

  FIG. 18 is a flowchart for explaining the control contents at the time of starting the fuel cell system of the present embodiment that is implemented by the controller 20. This start-up control is a control that is performed during start-up from a no-load state in which oxidant gas (air) is present in the anode 1a, the hydrogen supply flow path 2a, and the bypass flow path 6.

  The oxidant gas present in the anode 1a when the fuel cell 1 is started is air that has entered from the outside while the fuel cell is stopped, air that has cross-leaked from the cathode 1b to the anode 1a through the electrolyte membrane, or the like. is there.

  Further, the state before the start-up control of the present embodiment is started (that is, when the fuel cell system is stopped) is a state in which all the components are stopped. The purge valve 8 is opened, the hydrogen supply control valve 5 is closed, and the air pressure regulating valve 9 is opened.

  First, in step S61, it is determined whether or not a trigger for determining activation of the fuel cell system is on. As the activation determination trigger, for example, it can be assumed that the main switch 15 is in an on state. If the trigger is on, the process proceeds to step S62, and if it is off, the control is terminated. In step S62, the purge valve 8 is closed, and in step S63, the compressor 10 is started. The compressor 10 is controlled to a rotational speed corresponding to the idle state described above.

  Next, the circulation pump 7 is started at step S64. Here, the rate of change (increase rate) in the number of revolutions of the circulation pump 7 from the stop state of the circulation pump 7 until reaching a predetermined number of revolutions, which will be described later, is controlled at a substantially maximum changeable rate. In step S65, it is determined whether the rotational speed of the circulation pump 7 exceeds the maximum rotational speed during normal operation that can be stably controlled and is controlled to a predetermined rotational speed. The predetermined rotation speed of the circulation pump 7 in this embodiment is set to a rotation speed larger than the maximum rotation speed in the normal operation that can be stably controlled according to the fifth embodiment. For this reason, the circulation amount of the mixed gas composed of hydrogen and oxygen in this embodiment can be made larger than the circulation amount in the fifth embodiment.

  An example of the operating conditions of the circulation pump 7 in the sixth embodiment is shown in FIG. At time t1, the circulation pump 7 is started and the rotational speed is increased at the maximum change rate. At time t2, the maximum rotational speed of the normal operation region is exceeded and the control enters an unstable region. Here, the normal operation area is an area in which the rotation speed can be stably controlled, and the maximum rotation speed in the fifth embodiment means the maximum rotation speed in the normal operation area. The rotational speed of the circulation pump 7 that has entered the unstable region is controlled to be a predetermined rotational speed, although it is not stable in terms of control. At this time, the discharge amount of the circulation pump 7 is larger than the maximum discharge amount in the normal region. Become.

  In subsequent step S66, the hydrogen supply control valve 5 is opened, the pressure of hydrogen in the hydrogen supply flow path is increased, and hydrogen supply to the anode is started. Further, in step S67, the opening of the air pressure regulating valve 9 is narrowed to increase the air pressure in the cathode. At this time, the pressure increase time for increasing the pressure of hydrogen and air in the atmospheric pressure to a predetermined pressure is controlled to be the shortest time.

  Next, in step S68, it is determined whether or not the elapsed time after the start of hydrogen supply has passed a predetermined time. Here, the predetermined time is set to a time when hydrogen is expected to reach the anode. When the predetermined time has elapsed, the process proceeds to S69, where current extraction from the fuel cell 1 is started. Here, a minute load current corresponding to the idle state is taken out.

  In subsequent step S70, it is determined whether or not the elapsed time from the start of hydrogen supply in step S66 has exceeded a predetermined time required for consumption of oxygen remaining in the anode. Here, the time during which the concentration of oxygen remaining in the anode is 6% or less is defined as a predetermined time. When the elapsed time exceeds the predetermined time, the process proceeds to step S71, where the rotational speed of the circulation pump 7 is reduced to the rotational speed corresponding to the aforementioned idle state, and the control is finished.

  By adopting such start-up control, in this embodiment, as in the fifth embodiment, hydrogen (exhaust anode gas) is introduced into the anode in a circulation amount that can be realized in the fuel cell system, so that the anode Since the remaining oxygen is consumed quickly in a short time by this combustion reaction, the oxidative deterioration of the cathode as represented by the equation (3) can be minimized.

  In addition, by supplying the hydrogen to the anode by operating the circulation pump 7 at a rotational speed higher than the maximum rotational speed under normal operating conditions that can be controlled stably, more hydrogen is supplied to the anode with easy control. In addition, since the remaining oxygen is rapidly consumed in a shorter time due to the combustion reaction at the anode, it is possible to further suppress the oxidative deterioration at the cathode as shown in the formula (3).

  Further, when starting up, when the rotational speed of the circulation pump 7 is increased to a predetermined rotational speed, the rotational speed is controlled to a substantially maximum change rate within a controllable range so that the rotational speed can be determined in a short time. The rotational speed can be increased.

  Further, by controlling the pressure increase time until the hydrogen pressure reaches a predetermined pressure to be the shortest controllable time, more fuel is supplied to the anode of the fuel cell 1 in a short time, and therefore remains. Oxygen and hydrogen are consumed quickly by causing a combustion reaction in the anode in a short time. Therefore, it is possible to suppress the oxidative deterioration of the cathode as shown in the expression (3).

  Further, since it is estimated that the oxygen concentration has been reduced to a predetermined reduction with the elapsed time from the start of hydrogen supply, it is not necessary to provide a concentration sensor, and the cost of the system can be reduced.

  FIG. 20 is a configuration diagram of the seventh embodiment.

Compared with the configuration of the fifth embodiment shown in FIG. 13, this embodiment eliminates the oxygen concentration sensor 31, and uses a concentration sensor 36 that detects the carbon dioxide concentration in the gas discharged from the cathode as an air discharge flow. The difference is that a second bypass flow path 32 is provided which is installed in the path 3b and bypasses a part of the bypass flow path 6 on the upstream side of the circulation pump 7. In the second bypass passage 32, an auxiliary circulation pump 33 and control valves 34 and 35 are installed upstream and downstream thereof. Here, the output of the concentration sensor 36 is input to the controller 20, and the auxiliary circulation pump 33 and the control valves 34 and 35 are controlled by the controller 20. In the present embodiment, the deterioration of the cathode is suppressed by increasing the circulation amount in cooperation with the two circulation pumps 7 and 33.
FIG. 21 is a flowchart for explaining the control contents at the time of starting the fuel cell system of the present embodiment, which is performed by the controller 20. This start-up control is a control that is performed during start-up from a no-load state in which oxidant gas (air) is present in the anode 1a, the hydrogen supply passage 2a, and the bypass passage 6.

  The oxidant gas present in the anode 1a when the fuel cell 1 is started is air that has entered from the outside while the fuel cell is stopped, air that has cross-leaked from the cathode 1b to the anode 1a through the electrolyte membrane, or the like. is there.

  Further, the state before the start-up control of the present embodiment is started (that is, when the fuel cell system is stopped) is a state in which all the components are stopped. The purge valve 8 is opened, the hydrogen supply control valve 5 is closed, and the air pressure regulating valve 9 is opened.

  First, in step S81, it is determined whether or not a trigger for determining activation of the fuel cell system is on. As a trigger, for example, it can be assumed that the main switch 15 is in an ON state. If the activation determination trigger is on, the process proceeds to step S82, and if it is off, the control is terminated. In step S82, the purge valve 8 is closed, and in step S83, the compressor 10 is operated. The compressor 10 is controlled to a rotational speed corresponding to the idle state described above.

  Next, the circulation pump 7 is operated at step S84. Here, the rate of change (increase rate) of the rotational speed of the circulation pump 7 is controlled at a substantially controllable maximum rate of change. In step S85, it is determined whether the rotation speed of the circulation pump 7 exceeds the controllable maximum rotation speed and is controlled to a predetermined rotation speed. If it is controlled, the process proceeds to step S86. The predetermined number of rotations of the circulation pump 7 in this embodiment is controlled in the same manner as in the sixth embodiment. For this reason, the discharge amount of the circulation pump 7 is larger than the discharge amount in the fifth embodiment.

  In subsequent step S86, the control valves 34 and 35 installed upstream and downstream of the auxiliary circulation pump 33 are opened, and in step S87, the auxiliary circulation pump 33 is operated. Here, the rate of change (increase rate) of the rotational speed of the auxiliary circulation pump 33 is controlled at a substantially controllable maximum rate of change, as with the circulation pump 7. The rotational speed of the auxiliary circulation pump 33 is also controlled to be a predetermined rotational speed that is higher than the maximum rotational speed at which stable control is possible, although the control stability is lacking as in the case of the circulation pump 7.

  In step S88, it is determined whether the rotational speed of the auxiliary circulation pump 33 exceeds the controllable maximum rotational speed and is controlled to a predetermined rotational speed. If it is controlled, the process proceeds to step S89. In step S89, the hydrogen supply control valve 5 is opened, the hydrogen pressure in the hydrogen supply flow path is increased, and hydrogen supply to the anode is started.

  Further, in step S90, the opening of the air pressure regulating valve 9 is throttled to increase the air pressure in the cathode. At this time, the pressure increase time for increasing the pressure of hydrogen and air in the atmospheric pressure to a predetermined pressure is controlled to be the shortest time.

  Next, in step S91, it is determined whether the elapsed time after the start of hydrogen supply has passed a predetermined time. Here, the predetermined time is set to a time when hydrogen is expected to reach the anode. When the predetermined time has elapsed, the process proceeds to S92, where current extraction from the fuel cell 1 is started. Here, a minute load current corresponding to the idle state is taken out.

  In the subsequent step S93, the detection value of the concentration sensor 36 is read to determine whether or not the carbon dioxide concentration has become equal to or lower than a predetermined concentration. When the concentration is lower than a predetermined concentration (for example, 1%), it is determined that the carbon dioxide concentration has been reduced to the lower limit, and the process proceeds to step S94. Immediately after supplying hydrogen to the anode in step S89, carbon dioxide is generated due to carbon corrosion of the cathode. Thereafter, when the oxygen in the hydrogen supply path is consumed, the carbon dioxide concentration decreases and converges to about 1%, which is the carbon dioxide concentration in the air.

  In step S94, the rotational speed of the circulation pump 7 is reduced to the rotational speed corresponding to the aforementioned idle state, and in step S95, the operation of the auxiliary circulation pump 33 is stopped. In step S96, the control valves 34 and 35 are closed to finish the control.

  FIG. 22 shows the pressure of the gas flowing through the branch between the hydrogen supply channel 2a and the bypass channel 6 (hereinafter referred to as point A; see FIG. 20) to the fuel cell 1 to the first and second bypass channels 6 and 32. It is a figure which shows the change of (static pressure).

  A gas flows from the point A toward the fuel cell 1, and at that time, a pressure loss occurs in the piping that forms the hydrogen supply channel 2 a, and a larger pressure loss occurs in the fuel cell 1. Thereafter, the gas flowing out of the fuel cell 1 flows into the second bypass flow path 32, and the pressure rises due to the lift of the auxiliary circulation pump 33 installed in the second bypass flow path 32. Thereafter, the pressure drops due to the pressure loss caused by the piping of the bypass flow path 6 and the second bypass flow path 32, but the pressure is again increased by the hydrogen circulation pump 7, and the pressure loss due to the piping is generated to return to the point A.

  Here, it is assumed that the auxiliary circulation pump 33 is installed not at the second bypass passage 32 but at the upstream side of the bypass passage 6 in the vicinity of the circulation pump 7. Here, the distance from the outlet of the fuel cell 1 of the circulation pump 33 is set so that the assumed position is farther than the position described in the embodiment. The pressure change in this case is indicated by a dotted line in FIG. As can be seen from this comparison, the pressure inside the fuel cell 1 varies depending on the position of the auxiliary circulation pump 33, and the internal pressure of the fuel cell 1 can be lowered if the circulation pump is provided immediately after the outlet of the fuel cell 1. . Similarly, the same effect can be expected even if the circulation pump 7 is arranged immediately after the outlet of the fuel cell 1.

  However, since only the circulation pump 7 is used during normal operation, the output performance of the fuel cell 1 is higher when the hydrogen pressure is higher. For this reason, it is desirable that the auxiliary circulation pump 33 that is used only at the time of start-up is arranged immediately after the outlet of the fuel cell 1, and the circulation pump 7 is arranged at a position far from the outlet. Thus, by providing the auxiliary circulation pump 33 immediately after the outlet of the fuel cell 1, hydrogen can be introduced into the fuel cell 1 in a shorter time.

In this embodiment, the following effects can be expected in addition to the effects described in the sixth embodiment.
In other words, in addition to the circulation pump 7 during normal operation, the auxiliary circulation pump 33 for activation is provided, thereby increasing the circulation amount at the time of activation and consuming oxygen in the anode in a short time due to the combustion reaction at the anode. The oxidation deterioration in the cathode as shown in the equation (3) can be suppressed.

  Further, the corrosion of the carbon support as oxidation deterioration occurs in the reaction of the above-described equation (3). Therefore, by measuring the carbon dioxide concentration in the gas discharged from the cathode with the concentration sensor 36, the circulation amount can be reduced. Judgment can be made.

  Furthermore, by providing the auxiliary circulation pump 33 in the vicinity of the fuel outlet of the fuel cell 1 in the bypass paths 6 and 33, the suction negative pressure action causes the vicinity of the anode of the fuel cell 1 as compared with the case where it is provided in the vicinity of the inlet of the fuel cell 1. And the pressure near the outlet of the fuel cell 1 is relatively low. Therefore, the hydrogen supply time to the fuel cell 1 can be shortened.

  FIG. 23 is a configuration diagram of the eighth embodiment.

  Compared with the configuration of the fifth embodiment shown in FIG. 13, the present embodiment is provided with a third bypass channel 37 that bypasses the circulation pump 7 of the bypass channel 6, and circulates in the third bypass channel 37. The difference is that an auxiliary circulation pump 38 having a lift higher than that of the pump 7 is provided. Further, the second bypass flow path 32 includes a control valve 39 downstream of the auxiliary circulation pump 33 and a three-way valve 40 at the upstream branch between the bypass flow path 6 and the third bypass flow path 37. The auxiliary circulation pump 38, the control valve 39 and the three-way valve 40 are controlled by the controller 20.

  FIG. 24 is a flowchart for explaining the control contents at the time of starting the fuel cell system of the present embodiment that is implemented by the controller 20. This start-up control is a control that is performed during start-up from a no-load state in which oxidant gas (air) is present in the anode 1a, the hydrogen supply flow path 2a, and the bypass flow path.

  The oxidant gas present in the anode 1a when the fuel cell 1 is started is air that has entered from the outside while the fuel cell is stopped, air that has cross-leaked from the cathode 1b to the anode 1a through the electrolyte membrane, or the like. is there.

  Further, the state before the start-up control of the present embodiment is started (that is, when the fuel cell system is stopped) is a state in which all the components are stopped. The purge valve 8 is opened, the hydrogen supply control valve 5 is closed, and the air pressure regulating valve 9 is opened.

  First, in step S101, it is determined whether or not a trigger for determining activation of the fuel cell system is on. As a trigger, for example, it can be assumed that the main switch 15 is in an ON state. If the activation determination trigger is on, the process proceeds to step S102, and if it is off, the control is terminated. In step S102, the purge valve 8 is closed, and in the subsequent step S103, the compressor 10 is started. The compressor 10 is controlled to a rotational speed corresponding to the idle state described above.

  Next, in step S <b> 104, the control valve 39 installed in the third bypass flow path 37 is opened, and the three-way valve 40 is controlled so that gas flows through the third bypass flow path 37. In the subsequent step S105, the auxiliary circulation pump 38 is activated. Here, the rate of change (increase rate) of the rotational speed of the auxiliary circulation pump 38 is controlled at a substantially controllable maximum rate of change. The rotational speed of the auxiliary circulation pump 37 is also controlled so as to be a predetermined rotational speed that is higher than the rotational speed capable of stable control, although the control stability is lacking as in the seventh embodiment. Is done. Further, as described above, since the head capacity of the auxiliary circulation pump is larger than that of the circulation pump 7, the circulation amount is larger than those in the fifth and sixth embodiments.

  In step S106, it is determined whether or not the rotational speed of the auxiliary circulation pump 37 exceeds the controllable maximum rotational speed and is controlled to a predetermined rotational speed. If it is controlled, the process proceeds to step S107. In step S107, the hydrogen supply control valve 5 is opened, the pressure of hydrogen in the hydrogen supply flow path is increased, and hydrogen supply to the anode is started. Further, in step S108, the opening of the air pressure regulating valve 9 is throttled to increase the air pressure in the cathode. At this time, the pressure increase time for increasing the pressure of hydrogen and air in the atmospheric pressure to a predetermined pressure is controlled to be the shortest time.

  In the subsequent step S109, an external load that consumes electric power (not shown) is used as charge consuming means similar to the fixed resistor, and a current flowing according to the electromotive force generated in the fuel cell 1 is extracted. That is, the electric current to be taken out by the external load is not defined, but the current that flows according to the electromotive force of the fuel cell 1 changes like a fixed resistance.

  Then, the process proceeds to S110, where it is determined whether or not the oxygen concentration detected by the concentration sensor 31 has become a predetermined concentration (for example, 6%) or less. Take out.

  Here, the power manager 13 is operated in the load current (output) extraction mode corresponding to the normal operation of the fuel cell 1. In S110, since hydrogen and air are supplied to the anode and the cathode, respectively, when using the charge consuming function, it is desirable to make a determination based on the oxygen concentration. The method of determining the predetermined density is as described in the fifth embodiment. Subsequently, the auxiliary circulation pump 37 is stopped in S112, the control valve 39 is closed in S113, and the three-way valve 40 is switched to control the gas to flow through the bypass passage 6. Finally, in S114, the hydrogen circulation pump 7 is started to operate at a rotational speed corresponding to the idle state, the start-up control is terminated, and the normal control is started.

  Therefore, in the present embodiment, in addition to the effects of the sixth embodiment, the lifting capacity (= circulating force) of the auxiliary circulation pump 38 for activation is larger than the lifting capacity of the circulation pump 7 used during normal operation. It becomes possible to temporarily increase the circulation amount at the time of startup. Therefore, since the residual oxygen is consumed quickly in a short time due to the combustion reaction at the anode, the oxidative deterioration of the cathode as shown in the equation (3) can be minimized.

  Furthermore, since the oxidation deterioration reaction of the cathode is suppressed when the oxygen concentration in the vicinity of the anode and in the hydrogen supply path approaches 0 after the introduction of hydrogen, the circulation rate can be reduced from the viewpoint of deterioration.

  In addition, by utilizing the charge consuming means, the reaction of the formula (3) can be suppressed, and the oxidative deterioration at the cathode at the start-up can be remarkably reduced.

  FIG. 25 is a configuration diagram of the ninth embodiment.

  Compared with the configuration of the fifth embodiment shown in FIG. 13, this embodiment is provided with a combustion catalyst 41 and a gas-liquid separator 42 for removing oxygen in the hydrogen supply channel 2 a instead of the oxygen concentration sensor 31. The configuration is different.

  FIG. 26 is a flowchart for explaining the control contents at the time of starting the fuel cell system of the present embodiment that is implemented by the controller 20. This start-up control is a control that is performed during start-up from a no-load state in which oxidant gas (air) is present in the anode 1a, the hydrogen supply flow path 2a, and the bypass flow path 6.

  The oxidant gas present in the anode 1a when the fuel cell 1 is started is air that has entered from the outside while the fuel cell is stopped, air that has cross-leaked from the cathode 1b to the anode 1a through the electrolyte membrane, or the like. is there.

  Further, the state before the start-up control of the present embodiment is started (that is, when the fuel cell system is stopped) is a state in which all the components are stopped. The purge valve 8 is opened, the hydrogen supply control valve 5 is closed, and the air pressure regulating valve 9 is opened.

  First, in step S121, it is determined whether or not a trigger for determining activation of the fuel cell system is on. As a trigger, for example, it can be assumed that the main switch 15 is in an ON state. If the trigger for activation determination is on, the process proceeds to step S122, and if it is off, the control is terminated. In step S122, the purge valve 8 is closed, and in step S123, the compressor 10 is started. The compressor 10 is controlled to a rotational speed corresponding to the load current in the idle state described above.

  Next, the circulation pump 7 is started in step S124. Here, the rotation speed of the circulation pump 7 is controlled to the maximum rotation speed within a stably controllable range. In step S125, the rotational speed of the circulation pump 7 is detected, and it is determined whether or not it has converged stably to the maximum rotational speed. If converged, the process proceeds to step S126, and if not converged, the determination is repeated until converged. In step S126, the opening degree of the air pressure regulating valve 9 is reduced, and the pressure inside the cathode is started. Subsequently, in step S127, the hydrogen supply control valve 5 is opened, and hydrogen supply to the anode of the fuel cell 1 is started.

  Next, in step S128, it is determined whether or not an elapsed time after the start of hydrogen supply has passed a predetermined time. Here, the predetermined time is set to a time when hydrogen is expected to reach the anode. When the predetermined time has elapsed, the process proceeds to S129, and extraction of current from the fuel cell 1 is started. Here, a minute load current corresponding to the idle state is taken out.

  Subsequently, in S130, it is determined whether or not the elapsed time from the start of hydrogen supply in step S127 exceeds a predetermined time taken for the oxygen remaining in the anode to be consumed by the combustion catalyst 41. Here, the time during which the concentration of oxygen remaining in the anode is a predetermined concentration, for example, 6% or less, is defined as a predetermined time. If the elapsed time exceeds the predetermined time, the process proceeds to step S131, where the rotational speed of the circulation pump 7 is reduced to the rotational speed corresponding to the aforementioned idle state, and the control is terminated.

  In the present embodiment having the above configuration, the following effects can be obtained in addition to the effects of the fifth embodiment.

  That is, by providing the combustion catalyst (oxygen concentration reducing means) 41 at the hydrogen inlet of the fuel cell 1 and downstream of the branch between the hydrogen supply channel 2 a and the bypass channel 6, the circulation pump 7 starts operating immediately. When the exhaust anode gas containing oxygen is introduced (during the first circulation cycle), the concentration of oxygen remaining in the circulation flow path system including the hydrogen supply flow path 2a, the gas discharge flow path 2b, and the bypass flow path 6 is the combustion catalyst 41. Is reduced. For this reason, the oxygen concentration in the exhaust anode gas supplied to the fuel cell 1 is lowered. Accordingly, the time for which the residual oxygen is consumed by the combustion reaction at the anode of the first embodiment is shortened, and thus the oxidative deterioration of the cathode is further suppressed.

  Further, by providing the catalytic combustion 41 for burning fuel such as hydrogen, the combustion reaction of hydrogen and oxygen is performed mainly by the combustion catalyst 41 instead of the anode. Therefore, the oxygen concentration can be quickly reduced in a shorter time.

  When the water remaining in the bypass channel 6 is supplied to the anode of the fuel cell 1 together with the exhaust anode gas, the water in the anode spreads to the cathode through the electrolyte membrane, and the water in the vicinity of the cathode increases. Therefore, the reaction for promoting the oxidative deterioration of the cathode as shown in the formula (3) proceeds. Therefore, by removing the water introduced to the anode by the gas-liquid separator 42 in advance, the oxidative deterioration of the cathode can be suppressed. Moreover, the fuel distribution to each single cell in the fuel cell 1 can be improved by removing the water remaining in the bypass flow path 6 in advance.

  In addition, since water is generated by the combustion reaction in the combustion catalyst 41, the combustion catalyst 41 is provided in the gas-liquid separator 42 or in the upstream portion thereof, so that the moisture generated in the combustion catalyst 41 is effective in the gas-liquid separator 42. Can be removed.

  FIG. 27 is a configuration diagram of the tenth embodiment.

  In the present embodiment, a first oxygen adsorbent 51 and a first gas-liquid separator 52 are installed between the oxygen concentration sensor 31 and the fuel cell 1 as compared with the configuration of the fifth embodiment shown in FIG. Furthermore, the second oxygen adsorbent 53 and the second gas-liquid separator 54 are also installed upstream of the circulation pump 7 in the bypass channel 6.

  FIG. 28 is a flowchart for explaining the contents of control performed by the controller 20 when the fuel cell system according to this embodiment is started. This start-up control is a control that is performed during start-up from a no-load state in which oxidant gas (air) is present in the anode 1a, the hydrogen supply flow path 2a, and the bypass flow path 6.

  The oxidant gas present in the anode 1a when the fuel cell 1 is started is air that has entered from the outside while the fuel cell is stopped, air that has cross-leaked from the cathode 1b to the anode 1a through the electrolyte membrane, or the like. is there.

  Further, the state before the start-up control of the present embodiment is started (that is, when the fuel cell system is stopped) is a state in which all the components are stopped. The purge valve 8 is opened, the hydrogen supply control valve 5 is closed, and the air pressure regulating valve 9 is opened.

  First, in step S141, it is determined whether or not a trigger for determining activation of the fuel cell system is on. As a trigger for the activation determination, for example, it can be assumed that the main switch 15 is in an on state. If the trigger is on, the process proceeds to step S142, and if it is off, the control is terminated. In step S142, the purge valve 8 is closed, and in step S143, the compressor 10 is started. The compressor 10 is controlled to a rotational speed corresponding to the idle state described above.

  Next, the circulation pump 7 is started in step S144. Here, the rotation speed of the circulation pump 7 is controlled to the maximum rotation speed within a stably controllable range. In step S145, the rotational speed of the circulation pump 7 is detected, and it is determined whether or not it has converged stably to the maximum rotational speed. If it has converged, the process proceeds to step S146. If it has not converged, the determination is repeated until it converges. In step S146, the opening degree of the air pressure regulating valve 9 is reduced, and the pressure increase in the cathode is started. Subsequently, in step S147, the hydrogen supply control valve 5 is opened, and hydrogen supply to the anode of the fuel cell 1 is started.

  Next, in step S148, it is determined whether the elapsed time after the start of hydrogen supply has passed a predetermined time. Here, the predetermined time is set to a time when hydrogen is expected to reach the anode. When the predetermined time has elapsed, the process proceeds to S149, and in step S149, an external load that consumes power (not shown) is used as a charge consuming function similar to a fixed resistor, and the current that flows according to the electromotive force generated in the fuel cell 1 is calculated. Take out. That is, the electric current to be taken out by the external load is not defined, but the current that flows according to the electromotive force of the fuel cell 1 changes like a fixed resistance. Subsequently, the process proceeds to S150, where it is determined whether or not the oxygen concentration detected by the concentration sensor 31 has become a predetermined concentration (for example, 6%) or less. Take out. Here, the power manager 13 is operated in a load current (output) extraction mode corresponding to control during normal operation. In S150, since hydrogen and air are supplied to the anode and the cathode, respectively, it is desirable to make a determination based on the oxygen concentration when using the charge consuming function. The method of determining the predetermined density is as described in the fifth embodiment.

  In step S151, current extraction from the fuel cell 1 is started. Here, a minute load current corresponding to the idle state is taken out, and in step S152, the rotational speed of the circulation pump 7 is decreased to the above-described idle state, and the control is ended.

  In this embodiment, in addition to the effects of the fifth embodiment, the following effects can be achieved.

  That is, by providing oxygen adsorbents (oxygen concentration reducing means) 51 and 53 in the circulation channel system through which hydrogen, residual oxygen, etc. circulate, the oxygen concentration in the gas flowing through the circulation channel system becomes low, so that the anode The oxygen concentration to be introduced is lowered. Therefore, since the time for which the residual oxygen is consumed by the combustion reaction at the anode is shortened, the oxidative deterioration of the cathode as shown in the formula (3) is remarkably suppressed. Further, as compared with the combustion catalyst of the ninth embodiment, it is possible to suppress acceleration of fuel cell system performance deterioration due to an increase in the cation concentration (contamination concentration) in the drain water due to elution of the catalyst metal.

  Further, by providing a first oxygen adsorbent (oxygen concentration reducing means) 51 at the downstream of the branch between the hydrogen supply flow path 2a and the bypass flow path 6 (on the fuel cell 1 side) at the inlet of the fuel cell 1, circulation is achieved. When hydrogen is introduced immediately after the pump 7 starts operating (during the first circulation cycle), the oxygen concentration is low before oxygen remaining in the circulation flow path system and introduced hydrogen are introduced into the anode of the fuel cell 1. Become. Accordingly, since the time for which the residual oxygen is consumed by the combustion reaction at the anode is shortened, the oxidative deterioration of the cathode is further suppressed.

  In addition, by providing the first and second gas-liquid separators 52 and 54 in the circulation flow path system in which hydrogen, residual oxygen, and the like circulate, moisture supplied to the anode of the fuel cell 1 is reduced. Therefore, it is possible to suppress the performance deterioration of the circulation pump 7 that is easily affected by moisture. Here, as compared with the ninth embodiment, the second gas-liquid separator 54 is provided immediately upstream of the circulation pump 7 to sufficiently remove moisture. Further, the second oxygen adsorbing material 53 is provided immediately upstream of the circulation pump 7 to sufficiently remove oxygen.

  FIG. 29 shows time-series changes in the control current value flowing through the circulation pump 7 when the second gas-liquid separator 54 is provided immediately upstream of the circulation pump 7 and when it is not provided. It can be seen that the control is more accurately performed in a state closer to the target control current in the case where the second gas-liquid separator 54 is not provided. This is considered to be a difference in the amount of water, and it can be seen that in the absence of the second gas-liquid separator 54, the transient current value increases because water is mixed into the circulation pump 7.

  Further, when moisture remaining in the circulation channel system is supplied to the anode of the fuel cell 1 together with hydrogen, the moisture in the anode spreads to the cathode through the electrolyte membrane, and the moisture in the vicinity of the cathode increases. Accordingly, the reaction for promoting the oxidative deterioration of the cathode proceeds (for example, the formula (3)). Therefore, the above-described oxidative deterioration can be suppressed by previously removing the water led to the anode by the first gas-liquid separator 52. Moreover, the fuel distribution to each single cell in the fuel cell 1 can be improved by removing the water remaining in the circulation channel system in advance.

  The present invention is not limited to the above-described embodiment, and it is obvious that various modifications can be made within the scope of the technical idea of the present invention.

It is a block diagram of a fuel cell system. It is a block diagram of a controller. It is a flowchart figure explaining the control content of this embodiment. It is a figure which shows the result of having measured the carbon dioxide generation | occurrence | production by the carbon corrosion of a cathode when the mixed gas of various hydrogen / air concentration ratios is introduce | transduced into the anode. It is a figure which shows the result of having investigated the hydrogen concentration or oxygen concentration of the waste gas discharged | emitted from the anode with respect to the hydrogen concentration of the mixed gas which the anode introduce | transduced. It is a flowchart figure explaining the control content of 2nd Embodiment. It is a block diagram of 3rd Embodiment. It is a flowchart figure explaining the control content of 3rd Embodiment. It is a block diagram of 4th Embodiment. It is a flowchart figure explaining the control content of 4th Embodiment. It is a state figure explaining the state of a fuel cell. It is a figure explaining the conventional subject. It is a block diagram of 5th Embodiment. It is a flowchart figure explaining the control content of 5th Embodiment. It is a figure which shows the relationship between the carbon dioxide discharge amount measured at the cathode exit, and the rotation speed of a circulation pump. It is a figure which shows the result of having measured the amount of carbon dioxide in the air discharged | emitted from a cathode. It is a block diagram of 6th Embodiment. It is a flowchart figure explaining the control content of 6th Embodiment. It is a figure which shows an example of the driving | running condition of a circulation pump. It is a block diagram of 7th Embodiment. It is a flowchart figure explaining the control content of 7th Embodiment. It is a figure which shows the change of the pressure (static pressure) of the gas which flows through a circulation flow path system. It is a block diagram of 8th Embodiment. It is a flowchart figure explaining the control content of 8th Embodiment. It is a block diagram of 9th Embodiment. It is a flowchart figure explaining the control content of 9th Embodiment. It is a block diagram of 10th Embodiment. It is a flowchart figure explaining the control content of 10th Embodiment. It is a figure which shows the time-sequential change of the control electric current value which flows through a circulation pump by the presence or absence of a gas-liquid separator immediately upstream of a circulation pump.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Fuel cell 2 Hydrogen tank 2a Hydrogen supply flow path 2b Gas discharge flow path 3 Hydrogen tank main valve 4 Pressure reducing valve 5 Hydrogen supply control valve 6 Bypass flow path 7 Circulation pump 8 Purge valve 9 Air pressure regulating valve 10 Compressor 11a First pressure sensor 11b Second pressure sensor 11c Third pressure sensor 11d Fourth pressure sensor 12 Hydrogen concentration sensor 13 Power manager 14 Voltage sensor 15 Main switch 16 Oxygen concentration sensor 17 Differential pressure sensor 18 Circulation amount control valve 20 Controller 31 Oxygen concentration sensor 32 Second Bypass flow path 33 Auxiliary circulation pump 34 Control valve 35 Control valve 36 Carbon dioxide concentration sensor 37 Third bypass flow path 38 Auxiliary circulation pump 39 Control valve 40 Three-way valve 41 Combustion catalyst 42 Gas-liquid separator 51 First oxygen adsorbent 52 First 1 gas-liquid separator 53 second oxygen adsorbent 54 second gas-liquid separator

Claims (33)

A fuel cell comprising a hydrogen supply channel for supplying hydrogen from the outside to the anode of the fuel cell, and a bypass channel connected to the hydrogen supply channel and circulating the exhausted anode gas discharged from the anode to the anode In the system,
A gas composition control means for controlling a gas composition of a mixed gas composed of the hydrogen supplied to the anode and the exhaust anode gas;
A gas discharge control means for controlling the discharge of the exhaust anode gas to the outside;
A controller for controlling the gas composition control means and the gas discharge control means;
The controller starts up when oxidant gas is present in the anode.
Closing the gas discharge control valve, circulating the exhaust anode gas to the anode through the bypass flow path,
The fuel cell system, wherein the gas composition control means is controlled so that the concentration of the gas constituting the mixed gas becomes a predetermined concentration.   2. The fuel cell system according to claim 1, wherein the gas composition control means includes a circulation pump that is disposed at least in the bypass flow path and controls a circulation amount of the exhaust anode gas to the anode. Comprising a hydrogen concentration detection means for detecting the hydrogen concentration in the mixed gas supplied to the anode,
2. The fuel cell system according to claim 1, wherein the controller controls the gas composition control means so that the detected hydrogen concentration is equal to or higher than a predetermined hydrogen concentration. An oxygen concentration detection means for detecting the oxygen concentration in the mixed gas supplied to the anode;
2. The fuel cell system according to claim 1, wherein the controller controls the gas composition control means so that the detected oxygen concentration is equal to or lower than a predetermined oxygen concentration. Comprising oxygen concentration detection means for detecting the oxygen concentration in the exhaust anode gas discharged from the anode,
2. The fuel cell system according to claim 1, wherein the controller controls the gas composition control means so that the detected oxygen concentration is equal to or lower than a predetermined oxygen concentration. Comprising carbon dioxide concentration detection means for detecting the carbon dioxide concentration in the exhaust cathode gas discharged from the cathode,
2. The fuel cell system according to claim 1, wherein the controller controls the gas composition control means so that the detected carbon dioxide concentration is equal to or lower than a predetermined carbon dioxide concentration. Voltage detecting means for detecting the voltage of the fuel cell;
2. The fuel cell system according to claim 1, wherein the controller controls the gas composition control means so that a detected voltage change amount is equal to or less than a predetermined voltage change amount. 3. A hydrogen supply control valve for controlling the amount of hydrogen supplied to the anode from the outside;
3. The fuel cell system according to claim 2, wherein the controller closes the gas discharge control valve, opens the hydrogen supply control valve after starting the circulation pump, and supplies hydrogen to the anode. A circulation amount control valve for controlling a circulation amount of the exhaust anode gas downstream of the circulation pump of the bypass flow path;
Oxygen concentration detecting means for detecting the oxygen concentration of the bypass flow path,
The fuel cell system according to claim 2, wherein the controller controls the circulation amount control valve so that the detected oxygen concentration is less than a predetermined oxygen concentration.   The controller closes the gas discharge control valve and the circulation amount control valve, operates the circulation pump, and increases the pressure between the circulation amount control valve and the circulation means of the bypass flow path. The fuel according to claim 9, wherein the hydrogen supply control valve is opened to supply hydrogen to the anode, and the circulation amount control valve is controlled so that the detected oxygen concentration is less than a predetermined oxygen concentration. Battery system. Pressure detecting means for detecting the pressure in the hydrogen supply flow path is provided upstream and downstream of the hydrogen supply control valve,
The controller calculates the amount of hydrogen supplied to the anode based on the detected pressure value, and the flow rate of exhaust anode gas circulated from the bypass flow path to the anode is 30% or less with respect to the calculated amount of hydrogen. 3. The fuel cell system according to claim 2, wherein the gas composition control means is controlled so as to become.   The fuel cell system according to claim 3, wherein the predetermined hydrogen concentration is 70%.   The fuel cell system according to claim 4, wherein the predetermined oxygen concentration is 6%. The gas composition control means is a circulation pump installed in the bypass flow path,
2. The fuel cell system according to claim 1, wherein the circulation pump circulates the exhaust anode gas at a controllable maximum rotation speed when the fuel cell is started.   15. The fuel cell system according to claim 14, wherein the circulation pump circulates the exhaust anode gas at a maximum rotational speed during normal operation of the fuel cell when the fuel cell is started.   15. The fuel cell system according to claim 14, wherein the circulation pump circulates the exhaust anode gas at a rotational speed exceeding a maximum rotational speed during normal operation of the fuel cell when the fuel cell is started.   The fuel cell system according to any one of claims 14 to 16, further comprising an auxiliary circulation pump as the gas composition control means in the bypass channel. The auxiliary circulation pump is installed upstream of the circulation pump in the exhaust anode gas flow direction,
18. The fuel cell system according to claim 17, wherein when the fuel cell is started, exhaust anode gas is circulated by cooperation of the circulation pump and the auxiliary circulation pump. The auxiliary circulation pump is installed in parallel with the circulation pump, and has a lifting capacity larger than the lifting capacity of the circulation pump,
18. The fuel cell system according to claim 17, wherein when the fuel cell is started, exhaust anode gas is circulated by the auxiliary circulation pump.   The fuel cell system according to claim 14, wherein a combustion catalyst for reducing an oxygen concentration in the exhaust anode gas is provided in a flow path through which the exhaust anode gas circulates.   21. The fuel cell system according to claim 20, further comprising a gas-liquid separator that removes moisture in the exhaust anode gas downstream of the combustion catalyst in the flow direction of the exhaust anode gas.   The fuel cell system according to claim 21, wherein the combustion catalyst and the gas-liquid separator are installed between a branch of the hydrogen supply channel and the bypass channel and the fuel cell.   The combustion catalyst and the gas-liquid separator are installed between a branch of the hydrogen supply channel and the bypass channel and the fuel cell, and in the bypass channel upstream of the circulation pump. The fuel cell system according to claim 21. An oxygen concentration detecting means for detecting the oxygen concentration in the exhaust anode gas is provided;
20. The circulating pump and the auxiliary circulating pump are set in an idle state when an oxygen concentration detected at the time of starting the fuel cell is equal to or lower than a predetermined oxygen concentration. Fuel cell system. A compressor for supplying an oxidant gas to the fuel cell;
Comprising carbon dioxide concentration detection means for detecting the carbon dioxide concentration in the exhaust cathode gas discharged from the cathode of the fuel cell;
When the fuel cell is started, the compressor is started, and when the carbon dioxide concentration in the exhaust cathode gas discharged from the cathode of the fuel cell is equal to or lower than a predetermined carbon dioxide concentration, the circulation pump and the auxiliary circulation pump are idled. The fuel cell system according to any one of claims 14 to 19, wherein the fuel cell system is in a state. Concentration detection means for detecting the oxygen concentration or hydrogen concentration in the exhaust anode gas is provided,
15. The circulation amount of the exhaust anode gas is reduced when the detected oxygen concentration is equal to or lower than a predetermined oxygen concentration or the detected hydrogen concentration is equal to or higher than the predetermined hydrogen concentration when the fuel cell is started. 20. The fuel cell system according to any one of 19. A concentration detecting means for detecting the carbon dioxide concentration in the exhaust cathode gas;
The fuel cell system according to any one of claims 14 to 19, wherein when the detected carbon dioxide concentration is equal to or lower than a predetermined carbon dioxide concentration, the circulation amount of the exhaust anode gas is reduced.   28. The fuel cell system according to any one of claims 14 to 27, wherein an oxygen concentration reducing means for reducing an oxygen concentration in the exhaust anode gas is provided in a flow path through which the exhaust anode gas circulates.   29. The fuel cell system according to claim 28, wherein the oxygen concentration reducing means is installed immediately upstream of the inlet of the fuel cell through which exhaust anode gas flows.   30. The fuel cell system according to claim 28 or 29, wherein the oxygen concentration reducing means is an oxidation catalyst.   31. The fuel cell system according to claim 30, further comprising a second oxygen concentration reducing means for reducing the oxygen concentration in the exhaust anode gas in a flow path through which the exhaust anode gas upstream of the circulation pump circulates.   32. The fuel cell system according to any one of claims 1 to 31, further comprising gas-liquid separation means for removing moisture in the exhaust anode gas in a flow path through which the exhaust anode gas circulates.   32. The fuel cell system according to claim 28, wherein the gas-liquid separation unit is installed downstream of the oxygen reduction unit.