Fuel cell system

Abstract

To provide a fuel cell system that can switch the drain valve to a closed state with an inexpensive configuration. [Solution] The fuel cell system includes a drain valve provided in the drain channel, a pressure sensor that detects the anode pressure, which is the pressure of the fuel gas in the fuel gas supply channel, a current detection unit that detects the output current, and a controller that controls the injector to repeatedly inject and not inject fuel gas according to the output current detected by the current detection unit, and switches the drain valve from an open state to a closed state according to the anode pressure detected by the pressure sensor. The controller switches the drain valve from an open state to a closed state based on the rate of decrease in the anode pressure detected by the pressure sensor when the injector does not inject fuel gas.

Description

Technical Field

[0001] The present invention relates to a fuel cell system.

Background Art

[0002] In recent years, in order to enable more people to access affordable, reliable, sustainable, and advanced energy, technological development has been carried out on fuel cells that contribute to energy efficiency. As a technology related to this type of fuel cell, conventionally, a fuel cell system is known that includes a water storage section in a circulation flow path of fuel gas for storing liquid water generated by power generation, and discharges the liquid water in the water storage section by opening a drain valve (see, for example, Patent Document 1). The system described in Patent Document 1 includes a water level sensor for detecting the water level of liquid water, and switches the drain valve from an open state to a closed state based on the detection signal of the water level sensor.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the fuel cell system described in Patent Document 1 requires a water level sensor because it refers to the detection value of the water level sensor when switching the drain valve to the closed state, which leads to an increase in cost.

Means for Solving the Problems

[0005] A fuel cell system according to one aspect of the present invention comprises a fuel cell that generates electricity when anode gas and cathode gas are supplied to it, an injector that injects the anode gas supplied to the fuel cell via an anode supply channel, a water reservoir provided in the anode gas discharge channel through which the anode gas discharged from the fuel cell passes and which stores liquid water generated by the power generation of the fuel cell, a drain valve provided in a drain channel connected to the water reservoir, a pressure detection unit that detects the pressure of the anode gas in the anode supply channel, a load detection unit that detects the power generation load, and a control unit that controls the injector to repeatedly inject and not inject anode gas according to the power generation load detected by the load detection unit, and switches the drain valve from an open position that opens the drain channel to a closed position that blocks the drain channel according to the pressure detected by the pressure detection unit. The control unit switches the drain valve from an open state to a closed state based on the rate of pressure drop detected by the pressure detection unit when the anode gas is not injected by the injector. [Effects of the Invention]

[0006] According to the present invention, the drain valve can be switched from an open state to a closed state with an inexpensive configuration that does not use a water level sensor. [Brief explanation of the drawing]

[0007] [Figure 1] A diagram showing a schematic configuration of a fuel cell system according to an embodiment of the present invention. [Figure 2] Figure 1 shows a cross-sectional view of the main part of the fuel cell stack included in the fuel cell system. [Figure 3] A block diagram showing the control configuration of a fuel cell system according to an embodiment of the present invention. [Figure 4] A time chart showing an example of operation by a fuel cell system according to an embodiment of the present invention. [Figure 5] A flowchart showing an example of the process executed by the controller in Figure 3. [Figure 6A] A diagram showing an example of the injector injection pattern in the second injection mode. [Figure 6B]A diagram showing an example of the injector injection pattern in the third injection mode. [Figure 7] A flowchart showing another example of the process performed by the controller in Figure 3. [Figure 8] A flowchart showing a modified example of Figure 5. [Figure 9] A time chart showing an example of the operation obtained from the flowchart in Figure 8. [Modes for carrying out the invention]

[0008] Embodiments of the present invention will be described below with reference to Figures 1 to 9. Figure 1 is a diagram showing a schematic configuration of a fuel cell system 100 according to an embodiment of the present invention. The fuel cell system 100 in Figure 1 is mounted on a vehicle (fuel cell vehicle), for example, and generates electricity to be supplied to a drive motor.

[0009] As shown in Figure 1, the fuel cell system 100 includes a fuel cell stack 1, a fuel gas supply / discharge unit 2 that supplies fuel gas (anode gas) to the fuel cell stack 1 and discharges fuel gas from the fuel cell stack 1, an oxidant gas supply / discharge unit 3 that supplies oxidant gas (cathode gas) to the fuel cell stack 1 and discharges oxidant gas from the fuel cell stack 1, and a cooling medium supply / discharge unit 4 that supplies a cooling medium to the fuel cell stack 1 and discharges the cooling medium from the fuel cell stack. The fuel gas is, for example, hydrogen. The oxidant gas is, for example, air containing oxygen. The cooling medium is, for example, water or a coolant liquid containing ethylene glycol or propylene glycol.

[0010] Figure 2 is a cross-sectional view of the main part of the fuel cell stack 1. As shown in Figure 2, the fuel cell stack 1 has a cell stack 110 formed by stacking a plurality of power generation cells 101. Each power generation cell 101 has an electrode assembly (UEA) 102 having a membrane electrode assembly (MEA) including an electrolyte membrane and an electrode, and separators 103 arranged alternately with the electrode assembly 102. The separator 103 integrally includes an anode separator 103a positioned facing one side of the electrode assembly 102 and a cathode separator 103b positioned facing the other side of the electrode assembly 102.

[0011] A cooling channel PAw is formed inside the separator 103, which is surrounded by the anode separator 103a and the cathode separator 103b, through which a cooling medium flows. The flow of the cooling medium cools the power generation surface of the power generation cell 101. The surface of the separator 103 facing the electrode assembly 102 is made uneven by press molding or the like to form a gas channel between it and the electrode assembly 102. Between the electrode assembly 102 and the anode separator 103a, an anode channel PAa is formed by a recess through which fuel gas flows. Between the electrode assembly 102 and the cathode separator 103b, a cathode channel PAc is formed by a recess through which oxidizer gas flows.

[0012] Figure 2 includes a cross-sectional view of the membrane electrode assembly 104 as a cross-sectional view of the electrode assembly 102. As shown in the detailed view of part A in Figure 2, the membrane electrode assembly 104 has an electrolyte membrane 105, an anode electrode 106 provided on one side of the electrolyte membrane 105, and a cathode electrode 107 provided on the other side of the electrolyte membrane 105. The electrolyte membrane 105 is, for example, a solid polymer electrolyte membrane, and a thin film of a water-containing perfluorosulfonic acid polymer can be used. Not limited to fluorine-based electrolyte membranes, hydrocarbon-based electrolyte membranes can also be used.

[0013] The anode electrode 106 is formed on one side of the electrolyte membrane 105 and has an electrode catalyst layer 106a that serves as the reaction field for the electrode reaction, and a gas diffusion layer 106b provided on the surface of the electrode catalyst layer 106a opposite to the electrolyte membrane 105 and supplies fuel gas by diffusion. The cathode electrode 107 is formed on the other side of the electrolyte membrane 105 and has an electrode catalyst layer 107a that serves as the reaction field for the electrode reaction, and a gas diffusion layer 107b provided on the surface of the electrode catalyst layer 107a opposite to the electrolyte membrane 105 and supplies oxidizing gas by diffusion. An intermediate layer (underlayer) may also be provided between the electrode catalyst layers 106a, 107a and the gas diffusion layers 106b, 107b.

[0014] The electrode catalyst layers 106a and 107a contain a catalytic metal that promotes the electrochemical reaction between hydrogen in the fuel gas and oxygen in the oxidizing gas, a proton-conducting electrolyte (such as an ionomer), and electron-conducting carbon particles. The gas diffusion layers 106b and 107b are composed of a gas-permeable conductive material, such as a porous carbon body.

[0015] At the anode electrode 106, fuel gas (hydrogen) supplied via the anode channel PAa is ionized by the action of a catalyst and moves to the cathode electrode side through the electrolyte membrane 105. The electrons generated at this time pass through an external circuit and are extracted as electrical energy. At the cathode electrode 107, oxidizing gas (oxygen) supplied via the cathode channel PAc reacts with hydrogen ions introduced from the anode electrode 106 and electrons that have moved from the anode electrode 106 to produce water. The produced water (called generated water) provides appropriate humidity to the electrolyte membrane 105, and excess water is discharged to the outside of the electrode assembly 102 along the gas flow. The generated water on the cathode side also flows to the anode side by reverse diffusion through the electrolyte membrane 105. Therefore, generated water is contained in both the fuel gas and the oxidizing gas. Condensed water is also contained in both the fuel gas and the oxidizing gas. The generated water and condensed water are collectively called liquid water.

[0016] As shown in FIG. 1, the fuel gas supply / discharge unit 2 includes a tank 21 in which fuel gas (anode gas) is stored, a fuel gas supply passage PA21 that guides the fuel gas in the tank 21 to the fuel gas inlet 21a of the fuel cell stack 1, and a fuel gas discharge passage PA22 through which the fuel gas (fuel exhaust gas) discharged from the fuel gas outlet 21b of the fuel cell stack 1 flows. An injector 22 and an ejector 23 are arranged in the fuel gas supply passage PA21. A gas-liquid separator 24 is connected to the fuel gas discharge passage PA22.

[0017] The injector 22 is composed of a plurality of electromagnetic injectors connected in parallel. More specifically, the injector 22 has a plurality (three in the figure) of large injectors 221 to 223 and a single small injector 224. The configurations of the plurality of large injectors 221 to 223 are equal to each other. The aperture diameter of each of the large injectors 221 to 223 is larger than the aperture diameter of the small injector. Therefore, the injection amount per injection of each of the large injectors 221 to 223 is larger than the injection amount per injection of the small injector 224, and the injection capacity of each of the large injectors 221 to 223 is higher than the injection capacity of the small injector 224. Note that the number of the large injectors 221 to 223 and the small injector 224 connected in parallel is not limited to those described above.

[0018] The injector 22 is driven in response to a drive command from a controller 60 (FIG. 3). By driving the injector 22, fuel gas is injected toward the ejector 23. The injector 22 functions as a gas supply unit that supplies fuel gas to the fuel cell stack 1. The ejector 23 has a nozzle portion, a suction portion, a confluence portion, and a diffuser portion. The fuel gas injected from the injector 22 passes through the nozzle portion with a small diameter and then flows into the diffuser portion through the confluence portion. The fuel gas that has passed through the ejector 23 is supplied to the fuel cell stack 1 through the fuel gas inlet 21a. The ejector 23 may be composed of a plurality of ejectors connected in parallel instead of a single ejector. For example, a large ejector connected to the large injectors 221 to 223 and a small EjectorThey may be provided in parallel to constitute the ejector 23.

[0019] The fuel gas discharged from the fuel gas outlet 21b, that is, the fuel exhaust gas (anode off-gas), is separated into fuel gas and water (liquid water) by the gas-liquid separator 24. A drain channel PA23 is connected to the gas-liquid separator 24, and an electromagnetic drain valve 25 is interposed in the drain channel PA23. The drain valve 25 is configured to be switchable between an open position and a closed position. The liquid water separated by the gas-liquid separator 24 is stored in the water storage section 24a of the gas-liquid separator 24. When the drain valve 25 is opened, the liquid water in the water storage section 24a is discharged to the outside through the drain channel PA23 and the drain valve 25. When the drain valve 25 is closed, the drain channel PA23 is blocked, and the outflow of the liquid water is prevented.

[0020] The fuel gas separated by the gas-liquid separator 24 is guided to the circulation channel PA24. The ejector 23 is connected to the circulation channel PA24. Although not shown, a purge channel is also connected to the circulation channel PA24. An electromagnetic purge valve that can be opened and closed is interposed in the purge channel, and the fuel gas in the circulation channel PA24 can be discharged to the outside through the purge channel and the purge valve. valve The fuel gas separated by the gas-liquid separator 24 is sucked into the ejector 23 through the circulation channel PA24 by the flow of the fuel gas injected from the injector 22. The sucked fuel gas merges with the fuel gas that has passed through the nozzle section of the ejector 23 at the confluence section of the ejector 23, and after being made into a uniform flow in the diffuser section of the ejector 23, it is supplied to the fuel cell stack 1 through the fuel gas inlet 21a.

[0021]

[0022] ​The oxidizer gas supply and discharge section 3 includes an electrically operated air pump 31 that generates high-pressure oxidizer gas (cathode gas), an oxidizer gas supply channel PA31 that guides the oxidizer gas generated by the air pump 31 to the oxidizer gas inlet 31a of the fuel cell stack 1, and an oxidizer gas discharge channel PA32 through which the oxidizer gas (oxidizer exhaust gas) discharged from the oxidizer gas outlet 31b of the fuel cell stack 1 flows. The air pump 31 can also be called a compressor.

[0023] A humidifier 32 is positioned across the oxidizer gas supply channel PA31 and the oxidizer gas discharge channel PA32. In the humidifier 32, humidity exchange occurs between the oxidizer gas and the oxidizer exhaust gas, and the moisture (water vapor) contained in the oxidizer exhaust gas in the oxidizer gas discharge channel PA32 humidifies the oxidizer gas in the oxidizer gas supply channel PA31.

[0024] The oxidizer gas supply and discharge section 3 further includes a bypass channel PA33. The bypass channel PA33 is connected to the oxidizer gas supply channel PA31 upstream of the humidifier 32 and the oxidizer gas discharge channel PA32 downstream of the humidifier 32. Through the bypass channel PA33, the oxidizer gas can be supplied while bypassing the humidifier 32 and the fuel cell stack 1.

[0025] In the oxidizer gas supply channel PA31, an adjustable electromagnetic control valve 33 is provided between the bypass channel PA33 and the humidifier 32. In the oxidizer gas discharge channel PA32, an adjustable electromagnetic control valve 34 is provided between the bypass channel PA33 and the humidifier 32. An adjustable electromagnetic control valve 35 is provided in the bypass channel PA33. By controlling the air pump 31 and the control valves 33-35, the supply amount and pressure of the oxidizer gas supplied to the fuel cell stack 1 can be adjusted. Furthermore, by controlling the control valves 33-35, the amount of oxidizer gas bypassing the fuel cell stack 1 can be adjusted.

[0026] The cooling medium supply and discharge section 4 includes a cooling device 41, a cooling medium supply channel PA41 connecting the cooling device 41 to the cooling medium inlet 41a of the fuel cell stack 1, and a cooling medium discharge channel PA42 connecting the cooling device 41 to the cooling medium outlet 41b of the fuel cell stack 1. Although not shown in the figures, the cooling device 41 includes a pump that pressurizes the cooling medium toward the fuel cell stack 1, a heat exchanger (radiator) that cools the cooling medium that has been heated after passing through the fuel cell stack 1, and a cooling fan that blows cooling air to the heat exchanger.

[0027] A pressure sensor 51 is connected to the fuel gas supply channel PA21 downstream of the ejector 23. The pressure sensor 51 detects the inlet pressure (anode pressure Pa) of the fuel gas supplied to the fuel cell stack 1. A pressure sensor 52 is connected to the oxidizer gas supply channel PA31 downstream of the humidifier 32. The pressure sensor 52 detects the inlet pressure (cathode pressure Pc) of the oxidizer gas supplied to the fuel cell stack 1.

[0028] A temperature sensor 54 is connected to the cooling medium supply channel PA41 to detect the temperature of the cooling medium (refrigerant inlet temperature). A temperature sensor 55 is connected to the cooling medium discharge channel PA42 to detect the temperature of the cooling medium (refrigerant outlet temperature). The refrigerant inlet temperature and refrigerant outlet temperature are collectively called the refrigerant temperature. The gas temperatures of the fuel gas and oxidizer gas can be estimated from the refrigerant temperature.

[0029] The fuel cell system 100 according to this embodiment is characterized by the control method of the fuel gas supply and discharge unit 2, particularly the control method when switching the drain valve 25 from the open position to the closed position. This point will be explained below. Figure 3 is a block diagram showing the main components of the fuel cell system 100 according to an embodiment of the present invention. As shown in Figure 3, the fuel cell system 100 includes a controller 60, a current detection unit 56 connected to the controller 60, a pressure sensor 51, an injector 22, and a drain valve 25.

[0030] The current detection unit 56 is composed of, for example, a current sensor that detects the output current of the fuel cell stack 1. The output current corresponds to the power generation load. The fuel cell system 100 is configured to calculate the amount of power generation required by the vehicle (required power generation amount) and to generate power according to the required power generation amount. For this reason, the current detection unit 56 may calculate the output current using the required power generation amount. In other words, instead of the current detection unit 56 detecting the output current with a sensor, the output current may be calculated using parameters that have a correlation with the output current. To put it another way, the current detection unit 56 may be provided inside the controller 60. Thus, detecting the output current by the current detection unit 56 also includes calculating the output current.

[0031] The controller 60 is a computer comprising a processing unit having a CPU, ROM, RAM, and peripheral circuits. Functionally, the controller 60 has an injector control unit 61 and a drain control unit 62. The controller 60 performs predetermined processing based on the output current I detected by the current detection unit 56 and the anode pressure Pa detected by the pressure sensor 51, and outputs control signals to the injector 22 and the drain valve 25.

[0032] First, let's explain the drainage control unit 62. The drainage control unit 62 has a drainage determination unit 620 that determines whether or not to open or close the drain valve 25. Based on the signal from the current detection unit 56, the drainage control unit 62 calculates the amount of generated water produced by power generation and estimates the amount of liquid water in the water storage section 24a of the gas-liquid separator 24, i.e., the water level. The water level in the water storage section 24a gradually rises due to power generation. The drainage determination unit 620 determines whether or not the water level in the water storage section 24a has risen to a predetermined drainage start position.

[0033] The drainage control unit 62 maintains the drain valve 25 in the closed position until the drainage determination unit 620 determines that the water level has risen to the drainage start position. When the drainage determination unit 620 determines that the water level has risen to the drainage start position, the drainage control unit 62 outputs a control signal to the drain valve 25, switching the drain valve 25 from the closed position to the open position. As a result, the liquid water in the water storage unit 24a is discharged, and the water level drops.

[0034] Subsequently, the drainage determination unit 620 determines whether the water level in the water storage unit 24a has dropped to the drainage completion position, that is, whether drainage is complete. The drainage control unit 62 maintains the drain valve 25 in the open position until the drainage determination unit 620 determines that drainage is complete. When the drainage determination unit 620 determines that drainage is complete, the drainage control unit 62 outputs a control signal to the drain valve 25, switching the drain valve 25 from the open position to the closed position.

[0035] Incidentally, since power generation also occurs during drainage, the water level fluctuates. For this reason, it is difficult to accurately determine whether drainage is complete or not based on the signal from the current detection unit 56, and there is a risk that the timing of closing the drain valve 25 will be delayed. If the closing of the drain valve 25 is delayed, some of the fuel gas will leak out through the drain valve 25, and the power generation efficiency will deteriorate. On the other hand, if a water level sensor is installed in the water storage unit 24a to determine the completion of drainage, it will lead to an increase in costs. Therefore, in this embodiment, in order to determine whether drainage is complete or not with accuracy and in an inexpensive configuration, the drainage determination unit 620 is configured as follows.

[0036] Figure 4 is a time chart illustrating the processing of the drainage determination unit 620. More specifically, Figure 4 is a time chart showing an example of the operation of the injector 22 and the drain valve 25 over time, and an example of the change in the anode pressure Pa detected by the pressure sensor 51 over time. The injector 22 repeatedly opens (on) and closes (off) at a predetermined cycle according to the power generation load, based on a command from the injector control unit 61. The anode pressure Pa increases when the injector 22 is turned on and decreases when it is turned off. The rate of increase (slope) of the anode pressure Pa is greater than the rate of decrease in the anode pressure Pa.

[0037] In other words, the anode pressure Pa increases each time the injector 22 is turned on at time points t0, t2, t4, t7, and t9, and decreases each time the injector 22 is turned off at time points t1, t3, t5, t8, and t10. The period T1 from time point t1 to t2, the period T2 from time point t3 to t4, the period T3 from time point t5 to t7, and the period T4 from time point t8 to t9 are pressure drop periods during which the anode pressure Pa decreases. The drain valve 25 is closed at the initial time point t0. At time point t6 between time points t5 and t7, if the drain determination unit 620 determines that the water level in the water storage unit 24a has risen to the drain start position, the drain valve 25 is opened.

[0038] The drainage determination unit 620 calculates the rate of decrease α (α1 to α4) of the anode pressure Pa during each pressure drop period T (T1 to T4) based on the signal from the pressure sensor 51, each time the injection of fuel gas by the injector 22 is completed. The rate of decrease α is the amount of decrease in anode pressure Pa per unit time and corresponds to the slope of the characteristic shown in Figure 4, which shows the change in anode pressure Pa. The rate of decrease α (slope) can be calculated, for example, by the least squares method. The rate of decrease α can also be calculated by dividing the amount of decrease in anode pressure Pa during the pressure drop period T by the pressure drop period T. The rate of decrease α3 during the pressure drop period T3 is divided into the rate of decrease α3a when the drain valve 25 is closed from time t5 to t6, and the rate of decrease α3b when the drain valve 25 is opened from time t6 to t7.

[0039] The drainage determination unit 620 calculates the average value of the anode pressure Pa decrease rate α each time it calculates the decrease rate α while the drain valve 25 remains closed. For example, at time t2, when the decrease rate α1 is calculated, the unit calculates the average value αm of the decrease rates α0 and α1 calculated up to time t2. At time t4, when the decrease rate α2 is calculated, the unit calculates the average value αm of the decrease rates α0, α1, and α2 calculated up to time t4.

[0040] When the drain valve 25 is switched to the open position at time t6, the drain determination unit 620 temporarily stores the average value αm of the drop rates α calculated up to time t6 as the drop rate of the anode pressure Pa when the drain valve 25 is in the closed state (closed drop rate αm1) in memory. In the example in Figure 4, the average value αm of the drop rates α0, α1, α2, and α3a for the period up to time t6 is stored in memory as the closed drop rate αm1. If the drain valve 25 is switched to the open position in the middle of the drop of the anode pressure Pa (pressure drop period T3), the drop rate α3a for that period T3 may be excluded when calculating the closed drop rate αm1. That is, the average value αm of the drop rates α0, α1, and α2 may be used as the closed drop rate αm1. When the drain determination unit 620 stores the closed drop rate αm1 in memory, it resets the average value αm to 0 in preparation for the next processing.

[0041] Subsequently, the drainage determination unit 620 calculates the rate of decrease α of the anode pressure Pa when the drain valve 25 is open (open rate of decrease αm2). That is, it calculates the open rate of decrease α3b at time t7 and the open rate of decrease α4 at time t9. Until the drainage of the water storage section 24a is complete, the change in the rate of decrease α of the anode pressure Pa is small. On the other hand, once the drainage of the water storage section 24a is complete, a portion of the fuel gas is discharged through the drain valve 25, so the rate of decrease α of the anode pressure Pa becomes larger.

[0042] Therefore, each time the open-time descent ratio αm2 is calculated at time points t7 and t9, the drainage determination unit 620 determines whether the deviation Δα between the open-time descent ratio αm2 and the closed-time descent ratio αm1 is greater than or equal to a predetermined value Δα1. In other words, the drainage determination unit 620 determines whether the magnitude of the open-time descent ratio αm2 (α3b, α4) is greater than or equal to a predetermined value Δα1 than the magnitude of the closed-time descent ratio αm1. In Figure 4, the deviation Δα between the open-time descent ratio αm2 and the closed-time descent ratio αm1 calculated at time point t7 is less than the predetermined value Δα1. On the other hand, the deviation Δα between the open-time descent ratio α4 and the closed-time descent ratio αm1 calculated at time point t9 is greater than or equal to the predetermined value Δα1.

[0043] The drainage determination unit 620 determines that drainage is complete when it determines that the deviation Δα at time t9 is greater than or equal to a predetermined value Δα1. When the drainage determination unit 620 determines that drainage is complete, the drainage control unit 62 outputs a control signal to the drain valve 25 and switches the drain valve 25 to the closed position. This prevents the fuel gas from being discharged to the outside. If the drain valve 25 is switched to the open position during the decrease of the anode pressure Pa (pressure decrease period T3), the decrease rate α3b during that period T3 may be excluded when calculating the decrease rate αm2 when open. Alternatively, the decrease rate αm1 when closed may be corrected by adding the error and margin of the pressure sensor 51, and the drainage determination unit 620 may determine whether drainage is complete or not by calculating the deviation Δα between the corrected decrease rate αm1 when closed and the decrease rate αm2 when open.

[0044] Figure 5 is a flowchart showing an example of the process performed by the drainage control unit 62 of the controller 60. The process shown in this flowchart is started each time the drain valve 25 is switched to the closed position. The operation shown in Figure 4 is obtained when the controller 60 executes the process shown in the flowchart of Figure 5. For convenience, the flowchart of Figure 5 will be explained below with reference to the parameters of Figure 4.

[0045] As shown in Figure 5, first, in step S1, the controller 60 calculates the rate of decrease α of the anode pressure Pa based on the signal from the pressure sensor 51 each time the fuel gas injection by the injector 22 is completed. Furthermore, the controller 60 calculates the average value αm of the rate of decrease α during the closed period of the drain valve 25. Next, in step S2, the controller 60 determines whether the conditions for opening the drain valve 25 have been met based on the signal from the current detection unit 56. The conditions for opening are met, for example, when the water level of the liquid water calculated by the signal from the current detection unit 56 rises to the drain start position. If the condition in step S2 is affirmative, the process proceeds to step S3; otherwise, it returns to step S1.

[0046] In step S3, the controller 60 stores in memory the closed-time drop rate αm1, which is the average value of the drop rates α calculated until the open condition is met. Next, in step S4, the controller 60 outputs a control signal to the drain valve 25, switching the drain valve 25 from the closed position to the open position. As a result, the liquid water in the water reservoir 24a is discharged to the outside through the drain valve 25. Next, in step S5, each time the fuel gas injection by the injector 22 is completed, the controller 60 calculates the open-time drop rate αm2, which is the drop rate α of the anode pressure Pa, based on the signal from the pressure sensor 51.

[0047] Next, in step S6, the controller 60 calculates the deviation Δα between the closed descent rate αm1 and the open descent rate αm2, and determines whether the deviation Δα is greater than or equal to a predetermined value Δα1. If the result in step S6 is positive, the process proceeds to step S7; otherwise, it returns to step S5. In step S7, the controller 60 outputs a control signal to the drain valve 25, switching the drain valve 25 from the open position to the closed position. This completes the processing of the flowchart in Figure 5.

[0048] In this embodiment, the rate of decrease α of the anode pressure Pa is calculated based on the value detected by the pressure sensor 51, so it is possible to determine whether or not drainage from the water storage section 24a is complete without using a water level sensor or the like. That is, when drainage is complete, the fuel gas is discharged to the outside through the drainage channel PA 23, so the rate of decrease α of the anode pressure Pa increases. Since the determination of whether or not drainage is complete is made taking this point into consideration, it is possible to accurately determine whether or not drainage is complete with an inexpensive configuration that does not use a water level sensor or the like.

[0049] Next, the injector control unit 61 shown in Figure 3 will be described. The injector control unit 61 calculates a target flow rate of fuel gas according to the required power generation amount and outputs a control signal to the injector 22 so that the fuel gas according to the target flow rate is supplied to the fuel cell stack 1. More specifically, the injector control unit 61 calculates the target injection period and target injection amount (duty cycle) of the injector 22 and controls the injector 22 to inject fuel gas according to the target injection period and target injection amount.

[0050] In this case, the injector control unit 61 first determines the injection mode according to the power generation load (required power generation amount) in the mode determination unit 610. The injection modes are classified into a first injection mode M1 in which only a single large injector 221 injects, a second injection mode M2 ​​in which a single large injector 221 and a small injector 224 inject, a third injection mode M3 in which a pair of large injectors 221 and 222 inject, and a fourth injection mode M4 in which all large injectors 221 to 223 inject. Note that the large injectors 22 used for injection in the first injection mode M1 and the second injection mode M2 ​​do not have to be a single injector; for example, the number of injectors 22 performing injection may be increased as the required amount of hydrogen increases, or a predetermined number of injectors 22 may be used.

[0051] Figure 6A shows an example of the injection pattern of the large injector 221 and the small injector 224 in the second injection mode M2. The horizontal axis of Figure 6A represents time, and the vertical axis (pulse height) represents the injection amount per unit time. As shown in Figure 6A, in the second injection mode M2, fuel gas is injected from the large injector 221 and the small injector 224, respectively, along a pulsed waveform with an injection period Tint corresponding to the power generation load. More specifically, the large injector 221 and the small injector 224 are turned on at the same timing, and their injection start times are the same. However, the valve opening time Ti1 of the large injector 221 is shorter than the valve opening time Ti2 of the small injector 224. Therefore, the injection timings of the large injector 221 and the small injector 224 are different. The injection timings of the large injector 221 and the small injector 224 may be offset so that they do not overlap. In other words, the injection from the large injector 221 may be performed first, followed by the injection from the small injector 224.

[0052] Figure 6B shows an example of the injection pattern of large injectors 221 and 222 in the third injection mode M3. As shown in Figure 6B, even in the third injection mode M3, fuel gas is injected from large injectors 221 and 222 in a pulsed waveform with an injection period Tint corresponding to the power generation load. More specifically, large injectors 221 and 222 are turned on at the same timing, and their injection start times are the same. Also, the valve opening time Ti1 of large injectors 221 and 222 is the same. Therefore, the injection timing of large injectors 221 and 222 coincides with each other.

[0053] Although not shown in the diagram, the injection pattern of the large injectors 221-223 in the fourth injection mode M4 is the same as that shown in Figure 6B. That is, in the fourth injection mode M4, the large injectors 221-223 are turned on at the same timing, and the injection start time is the same. Also, the valve opening time Ti1 of the large injectors 221-223 is the same. Therefore, the injection timing of the large injectors 221-223 coincides with each other.

[0054] Here, within the injection cycle Tint, the time during which any of the multiple injectors 221 to 224 is injecting fuel gas is defined as the injection time, and the time during which none of the multiple injectors 221 to 224 are injecting fuel gas is defined as the non-injection time. As shown in Figure 6A, the injection time of the second injection mode M2 ​​is equal to the valve opening time Ti2 of the small injector 224, and the non-injection time is the time Ts2 obtained by subtracting the valve opening time Ti2 from the injection cycle Tint.

[0055] On the other hand, as shown in Figure 6B, the injection time in the third injection mode M3 is equal to the valve opening time Ti1 of the large injectors 221 and 222, and the non-injection time is the time Ts1 obtained by subtracting the valve opening time Ti1 from the injection period Tint. Similarly, in the fourth injection mode M4, the non-injection time is the time Ts1 obtained by subtracting the valve opening time Ti1 from the injection period Tint. Thus, in the third injection mode M3 and the fourth injection mode M4, the injection timings of the multiple large injectors 221, 222 and large injectors 221 to 223 coincide with each other. For this reason, the non-injection time Ts1 in the third injection mode M3 and the fourth injection mode M4 is longer than the non-injection time Ts2 in the second injection mode M2.

[0056] In this embodiment, as described above, the controller 60 calculates the rate of decrease α of the anode pressure Pa each time the injection of fuel gas by the injector 22 is completed, and determines whether or not drainage is complete based on the rate of decrease α. Therefore, in order to accurately determine whether or not drainage is complete, it is necessary to accurately calculate the rate of decrease α, and for this purpose, a longer non-injection time is preferable. Taking this into consideration, the mode determination unit 610 determines the injection mode so as to preferentially adopt the third injection mode M3 or the fourth injection mode M4 over the second injection mode M2.

[0057] Figure 7 is a flowchart showing an example of a process performed by the injector control unit 61 of the controller 60. The process shown in this flowchart is started, for example, when the change in the power generation load becomes small and a steady state is reached, and is repeated at a predetermined cycle while the steady state continues.

[0058] As shown in Figure 7, first, in step S11, the controller 60 calculates a target fuel gas flow rate corresponding to the output current I detected by the current detection unit 56. Next, in step S12, the controller determines whether the output current I is greater than or equal to a predetermined value I1. This determination determines whether the power generation load is low load or not, and the predetermined value I1 is set to a value that distinguishes between low load and medium load. If the result in step S12 is positive, the process proceeds to step S13; otherwise, it proceeds to step S17.

[0059] In step S17, the controller 60 determines the injection mode to either the first injection mode M1 or the second injection mode M2. For example, if a predetermined value I3 is set to be smaller than a predetermined value I1, and the output current I is less than or equal to the predetermined value I3, the controller 60 determines the injection mode to be the first injection mode M1, and if it is greater than or equal to the predetermined value I3 and less than the predetermined value I1, the controller 60 determines the injection mode to be the second injection mode M2. The predetermined value I3 is set to a value that distinguishes between low load and extremely low load. Therefore, the controller 60 determines the injection mode to be the first injection mode M1 when there is an extremely low load, and to the second injection mode M2 ​​when there is a low load. The injector 22 can inject in the second injection mode M2 ​​even when the output current I is greater than the predetermined value I1, but in this embodiment, the predetermined value I1 is set lower from the viewpoint of prioritizing the third injection mode M3 and the fourth injection mode M4 over the second injection mode M2 ​​as described above.

[0060] In step S13, the controller 60 determines whether the output current I detected by the current detection unit 56 is greater than or equal to a predetermined value I2. This determination determines whether the power generation load is a medium load. The predetermined value I2 is set to be greater than the predetermined value I1 and to distinguish between a medium load and a high load. If the result in step S13 is positive, the process proceeds to step S15. In step S15, the controller 60 determines the injection mode to the fourth injection mode M4. The predetermined value I1 may be a value that distinguishes between a low load and an extremely low load, and the predetermined value I2 may be a value that distinguishes between a low load and a medium load.

[0061] On the other hand, if step S13 is rejected, the process proceeds to step S14. In step S14, the controller 60 controls a pair of large injectors 221,222 Then, it is determined whether or not it is possible to supply the target flow rate of fuel gas according to the output current I. Step S14 is a process that is executed when the output current I is less than or equal to a predetermined value I2, so it is usually affirmed. However, if the target flow rate of fuel gas is the injector 221,222 The supply capacity may be exceeded, in which case step S14 is rejected. If step S14 is affirmed, proceed to step S16; otherwise, proceed to step S15. In step S16, the controller 60 determines the injection mode to the third injection mode.

[0062] The processes described in steps S11 to S17 are performed by the mode determination unit 610. Once the injection mode is determined in steps S15 to S17, the process proceeds to step S18. In step S18, the controller 60 calculates the target injection period Tint according to the output current I detected by the current detection unit 56. The target injection period Tint is set to a shorter value as the output current I increases. The target injection period Tint may also take into account the anode pressure Pa, and may be set to be shorter as the anode pressure Pa (more precisely, the partial pressure of hydrogen considering the hydrogen concentration) decreases.

[0063] The controller 60 can determine the target injection period Tint using a predetermined map or by a calculation formula. The predetermined map is, for example, a map that defines the relationship between the target flow rate and the target injection period Tint such that the target injection period Tint decreases sharply in the region where the target fuel gas flow rate is low, and decreases gradually in the region where the target flow rate is high. When the target flow rate is above a predetermined value, the target injection period Tint may be set to a certain lower limit.

[0064] Next, in step S19, the controller 60 calculates the target injection amount (referred to as the unit injection amount) for each injection cycle Tint of the multiple injectors 221 to 224, according to the injection mode. In this case, first, the controller 60 calculates the target injection amount (referred to as the total injection amount) for the injection cycle Tint when a single injector 22 is used. Next, the controller 60 calculates the unit injection amount by dividing the total injection amount by 2 in the third injection mode, and by dividing the total injection amount by 3 in the fourth injection mode. In the first injection mode, the controller 60 calculates the unit injection amount using the total injection amount as is.

[0065] In the second injection mode, the controller 60 determines the injection ratio of the large injector 221 and the small injector 224 according to a predetermined map and calculates the unit injection amount for each. The predetermined map is, for example, one in which the injection ratio of the large injector 221 increases as the total injection amount increases. The unit injection amount is proportional to the valve opening time Ti of the injector 22. Therefore, the controller 60 calculates the target valve opening time Ti for each injector 221 to 224 according to the unit injection amount.

[0066] Next, in step S20, the controller 60 outputs a control signal to the injector 22 so that each injector 221 to 224 opens for a target opening time Ti at a target injection period Tint.

[0067] In the above, the controller 60 determines whether the drainage of the liquid water stored in the water storage section 24a is complete by comparing the rate of decrease α of the anode pressure Pa when the drain valve 25 is closed and when it is open. However, the controller 60 may also determine whether the drainage is complete based on the change in the rate of decrease α of the anode pressure Pa when the drain valve 25 is open. Figure 8 is a flowchart showing an example of the process performed by the controller 60 in that case. Figure 8 is a modified version of Figure 5, and as with Figure 5, the process shown in the flowchart of Figure 8 is started each time the drain valve 25 is switched to the closed position. The same reference numerals are used for the same processes as in Figure 5.

[0068] In Figure 8, when the drain valve 25 is switched to the closed position, it is not necessary to calculate the average value of the anode pressure Pa drop rate α. Therefore, the process in step S1 is omitted, and the process starts from step S2. As shown in Figure 8, in step S2, the controller 60 determines whether the condition for opening the drain valve 25 has been met based on the signal from the current detection unit 56. Step S2 is repeated until it is confirmed. If it is confirmed in step S2, the process proceeds to step S4, where the controller 60 outputs a control signal to the drain valve 25, switching the drain valve 25 from the closed position to the open position.

[0069] Next, in step S8, the controller 60 determines, based on the signal from the pressure sensor 51, whether the change in the rate of decrease α of the anode pressure Pa is greater than or equal to a predetermined value. Figure 9 is a time chart showing an example of the operation obtained by the process in Figure 8. As shown in Figure 9, when the injector 22 is turned off at time t20, the anode pressure Pa decreases. At this time, if the drainage of the liquid water stored in the water reservoir 24a has not been completed, the rate of decrease α when fuel gas is not injected remains constant or nearly constant even after time t21, as shown by the dotted line. On the other hand, once drainage is completed, the rate of decrease α at time t21 becomes greater than or equal to a predetermined value, as shown by the solid line. Therefore, in step S6, the controller 60 determines whether the rate of decrease α of the anode pressure Pa has become greater than or equal to a predetermined value.

[0070] Step S8 is repeated until it is affirmed. If it is affirmed in step S8, the process proceeds to step S7, where the controller 60 outputs a control signal to the drain valve 25, switching the drain valve 25 from the open position to the closed position (see Figure 9). This completes the process of the flowchart in Figure 8.

[0071] This embodiment can provide the following effects and advantages. (1) The fuel cell system 100 includes a fuel cell stack 1 (fuel cell) that generates electricity when fuel gas (anode gas) and oxidant gas (cathode gas) are supplied to it, an injector 22 that injects fuel gas supplied to the fuel cell stack 1 via a fuel gas supply channel PA21 (anode gas supply channel), a water storage section 24a of a gas-liquid separator 24 provided in the fuel gas discharge channel PA22 (anode gas discharge channel) through which the fuel gas discharged from the fuel cell stack 1 passes, and which stores liquid water generated by the power generation of the fuel cell stack 1, and a drainage channel PA23 connected to the water storage section 24a. The system includes a drain valve 25, a pressure sensor 51 that detects the anode pressure Pa, which is the pressure of the fuel gas in the fuel gas supply passage PA21, a current detection unit 56 that detects the output current as a power generation load, and a controller 60 that controls the injector 22 to alternately inject and not inject fuel gas according to the output current detected by the current detection unit 56 (duty cycle control), and switches the drain valve 25 from an open position (open state) that opens the drain passage PA23 to a closed position (closed state) that blocks the drain passage PA23 according to the anode pressure Pa detected by the pressure sensor 51 (Figures 1 and 3). The controller 60 switches the drain valve 25 from the open position to the closed position based on the rate of decrease α of the anode pressure Pa detected by the pressure sensor 51 when the injector 22 does not inject fuel gas (Figures 4 and 5).

[0072] In this way, the drain valve 25 is switched from the open position to the closed position based on the rate of decrease α of the anode pressure Pa detected by the pressure sensor 51. This allows for an inexpensive configuration that does not use a water level sensor or the like, and enables accurate determination of whether or not the drainage of the liquid water stored in the water storage section 24a is complete. This allows the drain valve 25 to be switched to the closed position at an appropriate timing.

[0073] (2) The controller 60 switches the drain valve 25 from the open position to the closed position based on the difference Δα between the closed position drop rate αm1, which is the rate of decrease α of the anode pressure Pa detected by the pressure sensor 51 when the drain valve 25 is switched to the closed position when the injector 22 is not injecting fuel gas, and the open position drop rate αm2, which is the rate of decrease α of the anode pressure Pa detected by the pressure sensor 51 when the drain valve 25 is switched to the open position when the injector 22 is not injecting fuel gas (Figure 5). When drainage is complete, the fuel gas is discharged to the outside through the drain passage PA23, so the rate of decrease α of the anode pressure Pa increases. Taking this into consideration, the controller 60 determines whether drainage is complete or not based on the difference Δα between the closed position drop rate αm1 and the open position drop rate αm2 of the anode pressure Pa, so that it is possible to determine whether drainage is complete or not with good accuracy and switch the drain valve 25 to the closed position without delay.

[0074] (3) When the deviation Δα becomes greater than or equal to a predetermined value Δα1, the controller 60 switches the drain valve 25 from the open position to the closed position (Figure 5). This allows for accurate determination of whether drainage is complete or not, and enables the drain valve 25 to be switched to the closed position at the optimal timing.

[0075] (4) The injector 22 has a plurality of injectors 221 to 224 (Figure 1). When the output current I detected by the current detection unit 56 is greater than or equal to a predetermined value I1, the controller 60 controls the injectors 221, 222 or 221 to 223 so that their injection timings coincide (Figures 6B, 7). This makes it possible to lengthen the non-injection time Ts1 of the injector 22 and to calculate the anode pressure Pa drop rate α accurately.

[0076] (5) When the output current I detected by the current detection unit 56 is less than a predetermined value I1, the controller 60 controls the injectors 221 and 224 so that their injection timings are mismatched (Figures 6A and 67). This makes it easy to generate power at a low load according to the required amount of power.

[0077] (6) The multiple injectors 221 to 224 include multiple large injectors 221 to 223 with equal injection capabilities and a small injector 224 with a smaller injection capability than the large injectors 221 to 223 (Figure 1). The controller 60 controls the injectors 22 so that when the output current I detected by the current detection unit 56 is greater than or equal to a predetermined value I1, the multiple large injectors 221, 222 or 221 to 223 inject fuel gas at equal injection timings, and when the output current I is less than the predetermined value I1, a single large injector 221 and a small injector 224 inject fuel gas at different injection timings (Figures 6A, 6B, 7). This makes it possible to increase the non-injection time Ts1 in predetermined injection modes M3 and M4 while generating power according to the required power generation amount.

[0078] (7) The multiple large injectors 221 to 223 include three large injectors 221 to 223 arranged in parallel with each other (Figure 1). The controller 60 controls the injectors 22 so that when the output current I detected by the current detection unit 56 is greater than or equal to a predetermined value I1 (first predetermined value), two large injectors 221 and 222 inject fuel gas (third injection mode M3), and when the output current I is greater than or equal to a predetermined value I2 (second predetermined value) which is greater than the predetermined value I1, all three large injectors 221 to 223 inject fuel gas (fourth injection mode M4) (Figure 7). As a result, when the output current I is greater than or equal to a predetermined value I1, even if fuel gas can be injected using the three large injectors 221-223 by the fourth injection mode M4, fuel gas is injected using the two large injectors 221 and 222 by the third injection mode M3. This reduces the frequency of use of the large injector 223 and improves durability.

[0079] (8) The controller 60 calculates a target fuel gas flow rate according to the output current I detected by the current detection unit 56. When the output current I is greater than or equal to a predetermined value I1 and less than a predetermined value I2, the controller controls the injectors 22 so that the two large injectors 221 and 222 inject fuel gas when the target flow rate is less than or equal to the amount that can be injected by the two large injectors 221 and 222, and the three large injectors 221 to 223 inject fuel gas when the target flow rate is greater than the amount that can be injected by the large injectors 221 and 222 (Figure 7). As a result, the third injection mode M3 is prioritized over the fourth injection mode M4, and the injectors 22 can be used efficiently while considering the durability of the injectors 22.

[0080] (9) The controller 60 may switch the drain valve 25 from the open position to the closed position when the change in the rate of decrease α of the anode pressure Pa detected by the pressure sensor 51 becomes greater than a predetermined value while the drain valve 25 is switched to the open position (Figure 8). This makes it possible to determine early whether drainage is complete and minimizes the outflow of fuel gas from the drain valve 25. In this case, the controller 60 may switch the drain valve 25 from the open position to the closed position when the rate of decrease α becomes greater than a predetermined value, rather than when the rate of decrease α changes.

[0081] The above embodiment can be modified into various forms. Several modifications are described below. In the above embodiment, the current detection unit 56 is configured to detect the power generation current using a current sensor, but the power generation load may be detected using other sensors or by calculation, and the configuration of the load detection unit is not limited to that described above. In the above embodiment, a pressure sensor 51 is provided in the fuel gas supply channel PA21, but the configuration of the pressure detection unit is not limited to that described above as long as it detects the pressure of the anode gas in the anode supply channel. In the above embodiment, the water stored in the water storage section 24a of the gas-liquid separator 24 is discharged via the drain valve 25, but the configuration of the water storage section can be anything as long as it is provided in the anode gas discharge channel and stores the liquid water generated by the power generation of the fuel cell.

[0082] In the above embodiment, the controller 60, acting as the control unit, controls the injector 22 to repeatedly inject and not inject fuel gas according to the power generation load, and also switches the drain valve 25 from the open position to the closed position according to the anode pressure Pa. More specifically, when the difference Δα between the closed-position drop rate αm1 (first drop rate) of the anode pressure Pa when the drain valve 25 is switched to the closed position and the open-position drop rate αm2 (second drop rate) of the anode pressure Pa when the drain valve 25 is switched to the open position becomes greater than or equal to a predetermined value Δα1, the drain valve 25 is switched from the open position to the closed position. However, the configuration of the control unit can be anything as long as the drain valve 25 is switched to the closed state based on the drop rate of the anode pressure Pa when the injector 22 is not injecting anode gas.

[0083] In the above embodiment, the injector 22 has a plurality of large injectors 221 to 223 arranged in parallel with each other, and in the third injection mode M3, two (first predetermined number) large injectors 221 and 222 inject fuel gas, and in the fourth injection mode M4, three (second predetermined number) large injectors 221 to 223 inject fuel gas. However, the first predetermined number and the second predetermined number are not limited to those described above, as long as the second predetermined number is greater than the first predetermined number. Therefore, the configuration of the injector 22 is not limited to those described above. For example, the injector 22 may be composed of a plurality of injectors with equal injection volume capacity. The injector 22 may also be composed of a single injector.

[0084] The above describes an example of applying the fuel cell system 100 to a fuel cell vehicle, but the fuel cell system of the present invention can also be applied to vehicles other than fuel cell vehicles.

[0085] The above description is merely an example, and the present invention is not limited by the embodiments and modifications described above, as long as the features of the present invention are not impaired. It is also possible to arbitrarily combine one or more of the above embodiments and modifications, and to combine modifications with each other. [Explanation of Symbols]

[0086] 1 Fuel cell stack, 22 Injector, 25 Drain valve, 51 Pressure sensor, 56 Current detection unit, 60 Controller, 61 Injector control unit, 62 Drain control unit, 100 Fuel cell system, 221-223 Large injector, 224 Small injector, PA21 Fuel gas supply channel, PA22 Fuel gas discharge channel, PA23 Drain channel, αm1 Drop rate when closed, αm2 Drop rate when open

Claims

1. A fuel cell that generates electricity by being supplied with anode gas and cathode gas, An injector that injects anode gas supplied to the fuel cell via an anode supply channel, A water reservoir is provided in the anode gas discharge channel through which the anode gas discharged from the fuel cell passes, and which stores the liquid water generated by the power generation of the fuel cell. A drain valve is provided in the drainage channel connected to the water storage section, A pressure detection unit for detecting the pressure of the anode gas in the anode supply channel, A load detection unit that detects the power generation load, The system includes a control unit that controls the injector to alternate between injecting and not injecting anode gas according to the power generation load, and a control unit that switches the drain valve from an open position that opens the drain passage to a closed position that blocks the drain passage according to the pressure detected by the pressure detection unit, The fuel cell system is characterized in that the control unit switches the drain valve from an open state to a closed state based on the pressure drop rate detected by the pressure detection unit when the anode gas is not being injected by the injector.

2. In the fuel cell system according to claim 1, A fuel cell system characterized in that the control unit switches the drain valve from the open position to the closed position based on the deviation between a first pressure drop rate, which is the pressure drop rate detected by the pressure detection unit when the drain valve is switched to the closed position when the anode gas is not being injected by the injector, and a second pressure drop rate, which is the pressure drop rate detected by the pressure detection unit when the drain valve is switched to the open position when the anode gas is not being injected by the injector.

3. In the fuel cell system according to claim 2, The fuel cell system is characterized in that the control unit switches the drain valve from the open position to the closed position when the deviation exceeds a predetermined value.

4. In the fuel cell system according to any one of claims 1 to 3, The injector has a plurality of injectors, The fuel cell system is characterized in that the control unit controls the injectors so that the injection timing of the plurality of injectors coincides with each other when the power generation load is greater than or equal to a predetermined value.

5. In the fuel cell system according to claim 4, The fuel cell system is characterized in that the control unit controls the injectors such that the injection timing of the plurality of injectors becomes mismatched when the power generation load is less than the predetermined value.

6. In the fuel cell system according to claim 5, The plurality of injectors comprises a plurality of large injectors having equal injection capabilities and a plurality of small injectors having less injection capabilities than the large injectors. The fuel cell system is characterized in that the control unit controls the injectors such that when the power generation load is greater than or equal to a predetermined value, the plurality of large injectors inject anode gas at equal injection timings, and when the power generation load is less than the predetermined value, at least one of the large injectors and the small injectors inject anode gas at different injection timings.

7. In the fuel cell system according to claim 4, The plurality of injectors include a second predetermined number of large injectors, which are greater than the first predetermined number, and are arranged in parallel with each other. The fuel cell system is characterized in that the control unit controls the injectors such that when the power generation load is greater than or equal to a first predetermined value, a first predetermined number of large injectors inject anode gas, and when the power generation load is greater than or equal to a second predetermined value which is greater than the first predetermined value, a second predetermined number of large injectors inject anode gas.

8. In the fuel cell system according to claim 7, The fuel cell system is characterized in that the control unit calculates a target flow rate of anode gas according to the power generation load, and controls the injectors such that when the target flow rate is greater than or equal to a first predetermined value and less than a second predetermined value, the first predetermined number of large injectors inject anode gas, and when the target flow rate is greater than the amount that can be injected, the second predetermined number of large injectors inject anode gas.

9. In the fuel cell system according to any one of claims 1 to 3, The fuel cell system is characterized in that, when the drain valve is switched to the open position, the control unit switches the drain valve from the open position to the closed position when the pressure drop rate or change in the pressure drop rate detected by the pressure detection unit becomes greater than a predetermined value.

Images