JP5791070B2 - Solid oxide fuel cell system - Google Patents

Description

  The present invention relates to a solid oxide fuel cell system including a cell stack that generates electricity by reacting a fuel gas and an oxidizing material gas.

  A solid oxide fuel cell (SOFC) has a fuel electrode on one side with a solid electrolyte conducting oxygen ions in between, an air electrode on the other side, and a fuel gas on the fuel electrode side. (Hydrogen, carbon monoxide, etc.) is supplied, and an oxidant gas (air, oxygen, etc.) is supplied to the oxygen electrode side. As a material for the solid electrolyte, zirconia doped with yttria is generally used, and power generation is performed by causing an electrochemical reaction between the fuel gas and the oxidizing gas at a high temperature of 700 to 1000 ° C. Solid oxide fuel cells are being developed as a promising power generation technology because they can generate power with particularly high power generation efficiency compared to other fuel cell systems and gas engines.

  In general, a solid oxide fuel cell has a high operating efficiency because of its high operating temperature. However, since the operating temperature is high, the power generation performance of the solid oxide fuel cell cell stack deteriorates. It becomes a problem. The deterioration of the power generation performance of the cell stack is the deterioration of the output voltage of the cell stack, which can be grasped by the change in the characteristics (hereinafter also referred to as “IV characteristics”) between the output current and the output voltage. It can be separated into a slope on the plot in the IV characteristic (that is, an increase in internal resistance) and a decrease in open circuit voltage (OCV) (so-called intercept) in the IV characteristic (that is, a decrease in shielding performance). it can. In the case of solid oxide fuel cells, the main cause is the increase in resistance over time of the materials that make up the internal electrical passages. Specifically, the materials that join the fuel cells and between the fuel cells operate. The electrical resistance increases while transitioning to a more thermally stable state in temperature and atmosphere, and internal cracks occur in the direction that crosses the electrical path due to repeated heat cycles, oxidation, reduction, etc. For example, the effective area as a current path is reduced. Any of these can be regarded as a significant deterioration as a voltage change as the output current increases, and appears as a change in gradient in the IV characteristic.

  In addition, in the case of a cell stack that is an assembly of fuel cells, gas sealability and the like may decrease in a part of the cell stack, and in such a case, the open circuit voltage itself is very low locally. Become. In such a case, the output current can be confirmed as a voltage drop in any region. In particular, if the output current is significantly degraded as a voltage change under zero or low current conditions, it can be grasped separately from the increase in internal resistance described above, and the open circuit voltage (OCV) in the IV characteristics. Appears as a drop in the so-called (a drop in the intercept).

  For this reason, a method for diagnosing a deterioration state of a cell stack in a fuel cell system has been proposed (see, for example, Patent Document 1). According to this deterioration diagnosis method, the output power of the power converter of the fuel cell system is changed to change the input current (that is, the output current of the cell stack). Detect the output voltage. Then, the IV characteristic of the cell stack is estimated based on the detected current and the detected voltage, and the slope of the estimated IV characteristic and the output voltage at the time of no load (so-called estimated intercept) are obtained. Is determined to be in a deteriorated state when the deterioration standard is reached.

JP-A-11-195423

  However, in the above-described degradation diagnosis method, the degradation is diagnosed by obtaining the slope of the estimated IV characteristic of the cell stack and the output voltage (so-called estimated intercept) at the time of no load. There is a problem that deterioration due to a drop cannot be accurately detected. For example, the slope of the estimated IV characteristic varies depending on the supply amount of the fuel gas and the oxidant gas, and this slope increases when the supply amount of the fuel gas and the oxidant gas is insufficient. In addition, the estimated intercept of the estimated IV characteristic varies due to a variation in the composition ratio between the fuel gas and the reformed water or a temperature variation in the reformer or the like. Accordingly, even if the slope and estimated intercept of the estimated IV characteristic of the cell stack are simply obtained, it is due to the deterioration of the cell stack, due to the supply amount of the fuel gas and the oxidizing gas, or the fuel gas and the reforming water. It is difficult to accurately determine whether it is caused by fluctuations in the composition ratio and so on, and it is impossible to accurately diagnose deterioration of the cell stack. In addition, since the deterioration state of the entire cell stack is diagnosed, the local deterioration state of the cell stack cannot be diagnosed.

  In general, when the cell voltage decreases due to deterioration of the cell stack, it is effective to perform control to reduce the control upper limit of the output current of the cell stack. In particular, when the slope of the IV characteristic plot increases (internal resistance increases), the heat of the deteriorated portion increases in proportion to the current value due to the progress of deterioration, and the deterioration further increases due to the temperature rise. In order to prevent this, suppression of the maximum output is effective.

  However, in the control to reduce the upper limit of the output current, the output voltage is lowered due to deterioration and the output current is lowered, that is, the maximum generated power of the system is actively lowered. Not.

On the other hand, if the intercept (open circuit voltage) of the IV characteristic due to the deterioration of the gas seal performance among the deterioration of the cell stack, the fuel utilization rate of the system is lowered rather than reducing the control upper limit of the output current. It is effective to add surplus fuel. In such a case, the power generation efficiency of the cell stack decreases, but such a fuel cell is often applied to a cogeneration system that recovers and uses exhaust heat, and surplus fuel gas is burned in the cogeneration system. In order to recover it as hot water, the decrease in power generation efficiency can be recovered as heat, and as a result, there is no significant reduction in energy saving when the entire system is considered. Therefore, if a decrease in gas sealability can be accurately detected, it is possible to take measures for extending the life suitable for this deterioration, and to take measures that do not significantly impair the energy saving performance of the entire system .
An object of the present invention is to provide a solid oxide fuel cell system capable of accurately detecting deterioration of a cell stack due to gas sealing properties.

  Another object of the present invention is to provide a solid oxide fuel cell system capable of ensuring a long life while preserving the power generation output as much as possible, giving priority to customer merits.

The solid oxide fuel cell system according to claim 1 of the present invention includes a reformer for reforming a raw fuel gas, a reformed fuel gas reformed by the reformer, and an oxidizing material. A cell stack having a plurality of fuel cells for generating power by oxidation and reduction, a fuel pump for supplying raw fuel gas to the reformer, and an oxidation for supplying an oxidant to the cell stack A solid oxide fuel cell system comprising a material blower and a control means for controlling the fuel pump and the oxidant blower,
A first voltage detecting means for detecting a voltage between the first parts of the cell stack; and a second voltage detecting means for detecting a voltage between the second parts of the cell stack, the control The means includes a deterioration determination mode switching means for switching to a deterioration determination mode for determining a deterioration state of the cell stack, a virtual open circuit voltage calculating means for calculating a virtual open circuit voltage, and a deterioration state of the cell stack. Including a deterioration determining means for determining,
When switched to the deterioration determination mode by the deterioration determination mode switching means, the control means performs insertion control to the side where the output current of the cell stack becomes lower, and the first voltage detection means An output voltage between the first parts of the cell stack is detected, and the second voltage detection means detects an output voltage between the second parts of the cell stack during insertion control, and the virtual open circuit The circuit voltage calculation means calculates a first virtual open circuit voltage based on the detection voltage of the first voltage detection means, and calculates a second virtual open circuit voltage based on the detection voltage of the second voltage detection means. The deterioration determining means determines the deterioration state of the cell stack based on the first and second virtual open circuit voltages.

In the solid oxide fuel cell system according to claim 2 of the present invention, voltage detecting means is provided corresponding to each of the three or more parts of the cell stack, and each voltage detecting means is a corresponding part. The deterioration determination unit of the control unit determines a deterioration state of the cell stack using a detection voltage of the voltage detection unit.

Furthermore, in the solid oxide fuel cell system according to claim 3 of the present invention, the control means further includes a mode switching means for switching the operation mode, and the deterioration determining means is determined to be in a deteriorated state. Then, the mode switching means switches to the operation in the deterioration mode, and in the operation in the deterioration mode, the control means controls the operation of the system so that the fuel utilization rate in the cell stack is lowered.

According to the solid oxide fuel cell system of the first aspect of the present invention, when the cell stack is in a power generation state where power generation is performed at a constant output current, the deterioration determination mode is switched to. Since the deterioration state of the stack is diagnosed, the deterioration state of the cell stack can be diagnosed without starting and stopping the system. In particular, in a solid oxide fuel cell system operating at a high temperature, it is not preferable to increase the frequency of starting and stopping from the viewpoint of durability and energy saving, and the number of times of starting and stopping is, for example, one month. Therefore, the frequency of deterioration diagnosis is reduced in the power generation state. However, the frequency of deterioration diagnosis is increased by performing it in the power generation state. So that there is no delay in response.

In this deterioration determination mode, during normal power generation (for example, after a predetermined time has elapsed since the start of taking out the current value corresponding to the maximum output), control is performed to pull the generated current to the lower side, and the first voltage The detecting means detects the output voltage between the first parts of the cell stack at the time of this insertion / removal control, and the second voltage detecting means detects the output voltage between the second parts of the cell stack at the time of this insertion / removal control. . And based on the detection voltage of the 1st and 2nd voltage detection means, the current-voltage characteristic (IV characteristic) between 1st and 2nd site | parts is acquired, and a virtual open circuit voltage calculating means is this current-voltage. The characteristics are used to calculate the first and second virtual open circuit voltages, and the deterioration of the cell stack is diagnosed based on the first and second virtual open circuit voltages. Thus, the deterioration state is diagnosed. Thus, it is possible to diagnose the local deterioration state of the cell stack without being affected by temperature fluctuations or fuel gas composition fluctuations even during power generation.
The open circuit voltage that is the basis of the idea of the virtual open circuit voltage has a substantially fixed voltage value when the gas seal property of the cell stack is good, but tends to decrease when the gas seal property is deteriorated. In general, the cell stack is configured by connecting several tens to several hundreds of fuel cells, and in such a cell stack, when an open circuit voltage between the entire cell stacks is detected, for example, one fuel cell Even if the open circuit voltage of the cell decreases, it is difficult to detect the deterioration, and it is also difficult to determine whether the decrease of the open circuit voltage is caused by temperature fluctuation or fuel gas composition fluctuation.
For this reason, the output voltage between two parts (ie, the first and second parts) of the cell stack is detected, and the virtual open circuit voltage of these parts is calculated using the current-voltage characteristics. In this way, the deterioration of the cell stack is diagnosed based on these virtual open circuit voltages. By diagnosing the deterioration state in this way, the cell stack is not affected by temperature fluctuations or fuel gas composition fluctuations. A local degradation state can be determined.
In this case, the unit detection voltage per fuel cell in the first and second parts is calculated, and the first and second virtual open circuit voltages (first) per cell are calculated using the unit detection voltage. And the second virtual unit open circuit voltage).

  In the degradation determination mode, the generation current insertion / reduction control does not necessarily have to be changed to zero, and may be reduced to, for example, about 30% of the maximum output current. The current-voltage characteristics between the first and second parts can be acquired. In addition, the insertion time from the maximum output current to the minimum current when drawing to the lower side may be within several tens of seconds, for example, and since the current value is restored to the original value after the insertion control, the energy saving performance is also impaired. The deterioration state of the cell stack can be diagnosed without any problem.

In the solid oxide fuel cell system according to claim 2 of the present invention, the voltage detection means is provided corresponding to each of the three or more portions of the cell stack. It is possible to accurately diagnose local deterioration of a part, and it is possible to diagnose local deterioration of a finer part by increasing the number of installed voltage detection means.

Furthermore, according to the solid oxide fuel cell system according to claim 3 of the present invention, when the deterioration determining means determines that the cell stack is locally deteriorated, the mode switching means sets the deterioration mode, Operation in the deterioration mode is performed. In this deterioration mode, the system is operated so that the fuel utilization rate in the cell stack decreases. The fuel utilization rate is a ratio of how much generated current is taken out with respect to the supply rate of valence electrons contained in the fuel gas (anode gas).

  In the solid oxide fuel cell system, not all of the fuel gas input is converted into electrical energy by the cell stack, but the surplus fuel gas is intentionally left, and this surplus fuel gas is burned and reformed. The heat is used for heat of reforming in the vessel (reforming catalyst), heat of vaporization of the reformed water into water vapor, and heat for maintaining the cell stack at the operating temperature. Increasing this fuel utilization rate can be expected to increase the power generation efficiency, but on the other hand, the internal resistance increases due to the lowering of the operating temperature, and there is a difference in fuel distribution among the fuel cells. There is a risk of poor supply to the cell. This fuel utilization rate is set to about 70 to 75%, and if the fuel utilization rate is kept constant in a state where the gas sealability is lowered, there is a possibility that the deterioration of the fuel cell rapidly proceeds. By switching to the deterioration mode and reducing the fuel utilization rate, for example, to 65 to 70%, when the battery deteriorates, life-extending operation can be continued while suppressing rapid fuel cell deterioration. In addition, reducing the fuel utilization rate increases the combustion heat amount of the surplus fuel gas, but when combined with the cogeneration system, this combustion heat amount can be stored as hot water. There is no big drop.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a simplified diagram schematically showing an example of a solid oxide fuel cell system to which the present invention is applied . The block diagram which shows the control system in the solid oxide fuel cell system of FIG. The flowchart which shows the flow of control by the control system of FIG. The figure which shows the relationship between the electric power generation time when performing degradation determination mode at the time of starting, an open circuit voltage, and a voltage difference. The figure which shows the relationship between the electric power generation time when performing degradation determination mode at the time of completion | finish of operation, an open circuit voltage, and those voltage differences. The block diagram which shows the control system in one Embodiment of the solid oxide fuel cell system according to this invention. 7 is a flowchart showing a flow of control by the control system of FIG. The figure which shows the relationship between an output current and a cell voltage at the time of switching to a degradation determination mode during electric power generation, and carrying out insertion control of the generated electric current. The figure which shows the relationship between electric power generation time when performing degradation determination mode during electric power generation, virtual open circuit voltage, and those voltage differences.

Hereinafter, with reference to the accompanying drawings, a description will be given of an embodiment of a solid oxide fuel cell system according to the reference embodiments and the present invention of the solid oxide fuel cell system applying the present invention.

[ Reference Embodiment of Solid Oxide Fuel Cell System ]
A reference embodiment of a solid oxide fuel cell system to which the present invention is applied will be described with reference to FIGS. In FIG. 1, the illustrated solid oxide fuel cell system 2 includes a reformer 4 and a cell stack 6. In the reformer 4, for example, a catalyst in which ruthenium is supported on alumina is used as a reforming catalyst, and the raw fuel gas is steam reformed by the reforming catalyst as described later. As the raw fuel gas, natural gas (for example, city gas) is used.

  The cell stack 6 is configured by stacking a plurality of solid oxide fuel cells 8 for generating power by an electrochemical reaction via current collecting members. The fuel cell 8 includes a solid electrolyte 10 that conducts oxygen ions, a fuel electrode (not shown) provided on one side of the solid electrolyte 10, and an oxygen electrode (not shown) provided on the other side of the solid electrolyte 10. For example, zirconia doped with yttria is used as the solid electrolyte 10.

  The introduction side of the fuel electrode side 12 of each fuel cell 8 of the cell stack 6 is connected to the reformer 4 via the reformed fuel gas supply line 16, and the oxygen electrode side 14 of each fuel cell 8 is connected to the reformer 4. The introduction side is connected to an air blower 20 (functioning as an oxidizing material blower) via an air supply line 18. The air blower 22 is supplied to the cell stack 6 by supplying air as an oxidizing material to the air electrode side 14 of the cell stack 6 (each fuel cell 8) through the air supply line 18 and controlling the number of revolutions thereof. The amount of air supplied is controlled.

  The reformer 4 is connected to a desulfurization device 24 via a fuel gas supply line 22, and this desulfurization device 24 is connected to a raw fuel gas supply source (not shown) (for example, embedded) via a fuel gas supply line 26. Pipe, fuel gas tank, etc.). A fuel pump 28 is disposed in the fuel gas supply line 26. The fuel pump 28 supplies the raw fuel gas to the desulfurization device 24 through the fuel gas supply line 26, and the desulfurizer 34 removes the sulfur component contained in the fuel gas, and the raw fuel gas from which the sulfur component has been removed is the fuel. The amount of raw fuel gas supplied to the reformer 4 is controlled by controlling the number of revolutions of the fuel pump 28 that is supplied to the reformer 4 through the gas supply line 22.

In this reference embodiment , one end side of the water vapor supply line 30 is connected to the fuel gas supply line 22, and the other end side is connected to the vaporizer 32. The vaporizer 32 is modified via the reforming water supply line 34. It is connected to a quality water supply source (not shown) (for example, a water tank, a water pipe, etc.). A reforming water pump 36 is disposed in the reforming water supply line 34. The reforming water pump 36 supplies the reforming water to the vaporizer 32 through the reforming water supply line 34, and the supply amount of the reforming water supplied to the vaporizer 32 is controlled by controlling the rotation speed. The The vaporizer 32 vaporizes the reformed water to generate water vapor, and the generated water vapor is supplied to the fuel gas supply line 22 through the water vapor supply line 30, and the raw fuel gas is supplied through the fuel gas supply line 22. At the same time, it is fed to the reformer 4.

  The reformer 4 steam-reforms the raw fuel gas with the steam from the vaporizer 32, and the reformed fuel gas subjected to the steam reforming passes through the reformed fuel gas supply line 16 to the cell stack 6 (fuel cell 8). To the fuel electrode side 12. The steam supply line 30 may be connected to the reformer 4 so that the steam 30 from the vaporizer 32 may be directly fed to the reformer 4, or the vaporizer 32 and the reformer 4. And the raw fuel gas from the desulfurization device 24 may be supplied to the reformer 4 through the vaporizer 32.

  In the cell stack 6, the reformed fuel gas from the reformer 4 is supplied to the fuel electrode side 12 of each fuel cell 8, and the air (oxidant) from the air blower 20 is supplied to the air electrode side of each fuel cell 8. 14, and power generation is performed by oxidizing the reformed fuel gas and reducing the oxidizing material in each fuel cell 8. The power generation output of the cell stack 6 is connected to the inverter 41 via the power generation output line 38, and after the DC generated power is converted into AC power by the inverter 41, a power load (not shown) (for example, in the home) Supplied to various electrical appliances).

  A combustion chamber 40 is provided on each discharge side of the fuel electrode side 12 and the oxygen electrode side 14 of the cell stack 6, and surplus fuel gas discharged from the fuel electrode side 12 of each cell stack 6 (each fuel cell 8) ( That is, fuel gas that has not been used for power generation in the cell stack 6 and air (including oxygen) discharged from the oxygen electrode side 14 are discharged into the combustion chamber 40 and burned. An exhaust gas exhaust line 42 is connected to the combustion chamber 40, and exhaust gas from the combustion chamber 40 is exhausted to the atmosphere through the exhaust gas exhaust line 42.

  When such a solid oxide fuel cell system 2 is applied to a cogeneration system, although not shown, a hot water storage device is provided in association with the fuel cell system 2 and the exhaust gas discharge line 42 is heated. An exchanger is provided. The hot water storage device includes a hot water storage tank for storing hot water and a circulation line for circulating water in the hot water storage tank through a heat exchanger, and a circulation pump is disposed in the circulation line. In such a cogeneration system, heat exchange between the exhaust gas discharged through the exhaust gas discharge line 42 and the water circulated through the circulation line is performed in a heat exchanger, and hot water heated by the heat exchange is stored in hot water. The heat stored in the tank and thus the heat in the exhaust gas is recovered as hot water. By performing heat recovery in this way, energy saving can be improved.

The solid oxide fuel cell system 2 is further configured as follows in order to diagnose local deterioration of the cell stack 6. Referring to FIG. 2 together with FIG. 1, in this reference embodiment , the first voltage detecting means 52 is provided corresponding to the first part (left part in FIG. 1) on one side of the cell stack 6, and the other side. Corresponding to the second part (right part in FIG. 1), the second voltage detecting means 54 is provided. The first voltage detection means 52 detects a voltage between the first parts of the cell stack 6 (first open circuit voltage described later), and the second voltage detection means 54 is between the second parts of the cell stack 6. A voltage (second open circuit voltage described later) is detected.

  Further, a temperature detection means 56 is provided in the vicinity of the cell stack 6, and a current detection means 58 is provided on the power generation output line 42. The temperature detection unit 56 detects the temperature of the cell stack 6, and the current detection unit 58 detects the current of the generated power output through the power generation output line 42. Detection signals from the first and second voltage detection means 52 and 54, the temperature detection means 56 and the current detection means 58 are sent to a control means 60 for controlling the entire system.

  The control means 60 is composed of, for example, a microprocessor, and includes an operation control means 62, a voltage difference calculation means 64, a deterioration determination means 66, a mode switching means 68, and a memory 70. The operation control means 62 controls the operation of various components of the system (such as the fuel pump 28, the reforming water pump 36, the air blower 20, and the inverter 38), and the voltage difference calculation means 64 detects the first voltage detection means 52. The voltage difference between the voltage (voltage between the first parts) and the detection voltage (voltage between the second parts) of the second voltage detection means 54 is calculated, and the calculated voltage difference is used to determine the deterioration state, By using the voltage difference in this way, it is possible to perform the deterioration determination described later of the cell stack 6 while substantially eliminating the influence of temperature fluctuations and raw fuel gas composition fluctuations.

  When the number of fuel cells 8 in the first part of the cell stack 6 is the same as the number of fuel cells 8 in the second part, the detection voltages of the first and second voltage detection means 52 and 54 are detected. Can be used as is, but when the number of fuel cells 8 in the first and second parts is different, the detection voltage in the first and second parts is divided by the number of those parts. The detection voltage per unit (unit detection voltage) is preferably used as a detection voltage calculated by the voltage difference calculation means 64. In this case, the control means 60 includes a unit voltage calculation means (not shown). The unit voltage calculation means calculates the unit detection voltage per fuel cell 8 by dividing the detection voltage of the first voltage detection means 52 by the number of fuel cells in the first part for the first part. For the second part, the unit detection voltage per fuel cell 8 is calculated by dividing the detection voltage of the second voltage detecting means 54 by the number of fuel cells in the second part. By calculating the detection voltage, it is possible to easily diagnose the deterioration state in two parts having different numbers of fuel cells.

  Further, the deterioration determination unit 66 compares the voltage difference calculated by the voltage difference calculation unit 64 with the set voltage value registered in the memory unit 70. When the deterioration state does not proceed in the cell stack 6, there is almost no voltage difference, and when the local deterioration state proceeds, the voltage difference tends to increase. For this reason, when the voltage difference of the voltage difference calculation unit 64 exceeds the set voltage value, the deterioration determination unit 66 determines that the deterioration state has progressed and generates a deterioration signal.

  Further, the mode switching means 68 generates a mode switching signal based on the deterioration signal of the deterioration determining means 66, and switches from the normal mode operation to the deterioration mode operation based on the mode switching signal, so that the solid oxide fuel The battery system 2 is operated and operated in the deterioration mode.

  Next, the operation in the deterioration determination mode will be described with reference to FIGS. When the solid oxide fuel cell system 2 is started, the operation shifts to the deterioration determination mode, and the operation of the deterioration determination mode shown in FIG. 3 is executed. In operation in this deterioration determination mode, the operation control means 62 operates the fuel pump 28, the reforming water pump 36, the air blower 20, and the inverter 41, and the raw fuel gas is supplied to the reformer 4 through the desulfurization device 24. At the same time, the reformed water is vaporized by the vaporizer 32 and then supplied to the reformer 4 as steam, and the reformer 4 performs steam reforming of the raw fuel gas using the steam. Then, the reformed fuel gas subjected to steam reforming is supplied to the fuel electrode side 12 of the cell stack 6 and air as an oxidant is supplied to the oxygen electrode side 14 of the cell stack 6. Power generation is started, and surplus fuel gas is combusted in the combustion chamber 40, and the cell stack 6, the vaporizer 32, and the reformer 4 are heated using this combustion heat. At this time, the temperature of the cell stack 6 is low and the power generation state in the cell stack 6 is unstable, and the inverter 41 does not convert the power generation output from the cell stack 6 into AC power and output it.

  When started in this way, the temperature detecting means 56 detects the temperature of the cell stack 6 (step S1), and when the detected temperature of the temperature detecting means 56 reaches a measured temperature (for example, set to about 480 ° C.), the step The process proceeds to step S3 via S2, and the first and second voltage detection means 52 and 54 detect the voltage between the first and second portions of the cell stack 6. At this time, the generated power is not output from the inverter 41. Therefore, the voltages detected by the first and second voltage detecting means 52 and 54 are the first and second voltages between the first and second parts of the cell stack 6. 2 Open circuit voltage.

  When the voltage between the first and second portions (that is, the first and second open circuit voltages) is detected in this way, the voltage difference calculation means 64 detects the detected voltages of the first and second voltage detection means 52 and 54. (Step S4), and if this voltage difference is smaller than the predetermined voltage value, the process proceeds to step S6 assuming that local degradation of the cell stack 6 has not occurred. In step S6, the control unit 60 sets the normal mode. When the temperature detected by the temperature detection unit 56 reaches the operating temperature (for example, set to about 650 ° C.), the power generation state in the cell stack 6 is stabilized. Proceeding from step S7 to step S8, the solid oxide fuel cell system 2 is operated in the normal mode. In the operation in the normal mode, the fuel utilization rate in the cell stack 6 is set to about 70 to 75%, and the operation control means 62 determines that the fuel utilization rate is within this range. And the air blower 20 is controlled to operate.

  If the voltage difference calculated by the voltage difference calculation means 64 (that is, the voltage difference between the first and second open circuit voltages) exceeds a predetermined voltage value, it is determined that local degradation has occurred in the cell stack 6 in step S5. From step S9, the deterioration determining means 66 determines that the cell stack 6 is in a deteriorated state, and generates a deterioration signal. Thus, the mode switching means 68 generates a mode switching signal based on this deterioration signal, and sets the deterioration mode based on this mode switching signal (step S10).

  Thereafter, when the temperature detected by the temperature detecting means 56 reaches the operating temperature (for example, set to about 650 ° C.), the power generation state in the cell stack 6 is stabilized, and the process proceeds from step S11 to step S12. The battery system 2 is operated in the deterioration mode. In the operation in the deterioration mode, the fuel utilization rate in the cell stack 6 is lower than that in the normal mode, for example, set to about 65 to 70%, and the operation control means 62 is configured so that the fuel utilization rate falls within this range. 28, operation control of the reforming water pump 36 and the air blower 20 is performed. For example, in order to reduce the fuel utilization rate, the rotational speed of the fuel pump 28 may be increased somewhat to increase the supply amount of the raw fuel gas, or in addition to the increase in the rotational speed of the fuel pump 28. The air supply amount may be increased by increasing the rotational speed of the air blower 20 somewhat.

  The solid oxide fuel cell system 2 shown in FIGS. 1 to 3 is used, and the voltage difference between the detection voltages (that is, the first and second open circuit voltages) of the first and second voltage detection means 52 and 54 is used. An endurance test was conducted to confirm whether the deterioration state could be diagnosed. In this test, the detection voltage per fuel cell (ie, the detection voltage of the first and second voltage detection means 52, 54 divided by the number of fuel cells 8 in the first and second parts (ie, The voltage difference is calculated using the unit detection voltage (which becomes the first and second unit open circuit voltages).

  In this confirmation test, when the temperature of the cell stack is in the range of 460 to 580 ° C., the voltage between the first and second parts of the cell stack 6 is detected by the first and second voltage detecting means 52 and 54. The voltage difference (voltage difference of unit detection voltage) is calculated, and the unit detection voltage (mV / cell) and the unit detection voltage difference (mV / cell) between the first and second parts are expressed as follows: As shown in FIG. As shown in FIG. 4, in the voltage difference of the unit detection voltage (first and second unit open circuit voltage) between the first and second parts, the composition ratio of the raw fuel gas and the reformed water is It is easily understood that the change caused by only the deterioration of the fuel cell 8 is captured by canceling the fluctuation and the temperature fluctuation of the reformer 4.

  In the initial state of the solid oxide fuel cell system 2, the voltage difference between unit detection voltages (right axis in FIG. 4) is −0.8 mV, and this voltage difference did not change significantly until 11,000 hours. However, the continuous decrease of the voltage difference became remarkable after 11,000 hours, and the rate of the decrease became abrupt. From this tendency, in order not to include erroneous detection due to measurement error, for example, if it is set so that −10 mV or less is regarded as a deteriorated state, the cell stack 6 deteriorates after 14,000 hours have passed. Therefore, it is possible to shift to the operation in the deterioration mode for protecting the fuel cell 2.

In the reference embodiment described above, when the voltage difference between the detection voltages (first and second open circuit voltages) of the first and second voltage detection means 52 and 54 reaches a predetermined value, local degradation occurs in the cell stack 6. However, for example, when a change of a predetermined width or more occurs continuously for a predetermined time, it may be determined that local deterioration has occurred in the cell stack 6, or this When the number of times that the voltage difference reaches a predetermined value becomes a predetermined number (for example, 2 to 3 times), it may be determined that the cell stack 6 has locally deteriorated.

  In such a case, referring to FIG. 4, the detected voltages of the first and second voltage detecting means 52, 54 at the time of starting the solid oxide fuel cell system 2 (in FIG. Voltage difference (shown as 2 unit open circuit voltage) (voltage difference between first and second unit open circuit voltage), for example, ± 5.0 mV / 1000 hours voltage difference change rate continues with two startups In this endurance test, it is determined that the cell stack 6 has deteriorated at the end of 13,300 hours, and at this time, the deterioration mode operation is started. Can be migrated.

In the above-described reference embodiment , the deterioration determination mode is performed when the solid oxide fuel cell system 2 is driven to diagnose the deterioration state of the cell stack 6. 2 may be performed at the end of operation. At the end of this operation, cooling is performed while sending fuel gas and air to the fuel electrode side 12 and the air electrode side 14 of the cell stack 6 (fuel cell 8) while the output current is zero. The first and second open circuit voltages between the first and second portions of the cell stack 6 can be detected by the first and second voltage detecting means 52 and 54.

  FIG. 5 shows the experimental results when the deterioration judgment mode is performed at the end of operation. Also in FIG. 5, the detection voltage per fuel cell (that is, the unit) is obtained by dividing the detection voltage of the first and second voltage detection means 52 and 54 by the number of fuel cells 8 in the first and second parts. The voltage difference is calculated using the detected voltage (which becomes the first and second unit open circuit voltages). FIG. 5 shows the first and second unit open circuit voltages between the first and second portions and the voltage difference between them after 40 minutes from the end of operation.

  As shown in FIG. 5, the voltage difference between the first and second unit open circuit voltages (right axis in FIG. 5) is a slight change in voltage difference (from −2.3 mV to − 6.6 mV), but this voltage difference is almost constant thereafter. And it turns out that the fall has become remarkable since about 6500 hours passed. By using the open circuit voltage at the end of the operation, the deterioration state can be determined earlier than when using the open circuit voltage at the start-up, and an early response to the operation in the deterioration mode is possible. I understand that.

  The reason why the open circuit voltage at the end of operation can detect the deterioration state with higher sensitivity is considered as follows. In the fuel electrode of the fuel battery cell 6 whose gas sealing performance has deteriorated, it is considered that a sufficient amount of reducing gas cannot be supplied in the power generation state, and a partial oxidation state occurs. Since the generated oxidation state remains, it is considered that an open circuit voltage slightly lower than the original electromotive force is generated. On the other hand, since the fuel gas is sufficiently supplied to the fuel electrode of the fuel cell 6 at the time of start-up, even if an oxidation state is generated in the fuel electrode 6 in the fuel cell 6 in which the gas sealing property is lowered, It is thought that it has been reduced again in the process of rising and has returned to its original electromotive force.

In this reference embodiment , the open circuit voltage between the first part on one side of the cell stack 6 and the second part on the other side is detected, but the part for detecting the open circuit voltage is different in this way. The second part (or the first part) may include the first part (or the second part). For example, the part on one side (or the other side) of the cell stack 6 may be the first part (or the second part). One part (or second part) may be used, and the entire cell stack 6 may be used as the second part (or first part). In this embodiment, an open circuit voltage between two parts of the cell stack 6 is detected. However, an open circuit voltage between three or more parts of the cell stack 6 may be detected. By increasing the interval, a local deterioration state of the cell stack 6 between smaller parts can be diagnosed.

[ One Embodiment of Solid Oxide Fuel Cell System ]
Next, with reference to FIGS. 6 and 7, a description will be given of an embodiment of a solid oxide fuel cell system according to the present invention. In this embodiment , instead of directly detecting the open circuit voltage between the first and second portions of the fuel cell stack, a current − obtained by controlling the generated current to be pulled to the lower side during power generation − A virtual open circuit voltage is obtained using voltage characteristics (IV characteristics), and a local deterioration state of the cell stack 6 is diagnosed using the virtual open circuit voltage. Note that in this embodiment shaped condition, the same reference numerals are given to those similar to the reference embodiment described above, description thereof will be omitted.

In this embodiment , the basic configuration of the solid oxide fuel cell system is the same as that of the reference embodiment described above , and has the configuration shown in FIG. 1, and the configuration of the control means 60A is different. .

In FIG. 6, the control means 60A in this embodiment includes an operation control means 62, a voltage difference calculation means 64, a deterioration determination means 66, a mode switching means 68 and a memory means 70A, in addition to a deterioration determination mode switching means 82 and a virtual open. Circuit voltage calculation means 84 is included. The deterioration determination mode switching means 82 is switched to the deterioration determination mode for diagnosing the deterioration state of the cell stack 6 during the operation in the normal mode, and the virtual open circuit voltage calculation means 84 is obtained in the operation in the deterioration determination mode. Based on the current-voltage characteristics, the virtual open circuit voltage is calculated as described later. Other configurations in this embodiment are substantially the same as those in the reference embodiment described above.

The operation in the deterioration determination mode in this embodiment is performed as follows. Referring also to FIG. 7, switching to the deterioration determination mode is performed during the operation of the solid oxide fuel cell system in the normal mode. Switching to such a deterioration determination mode is such that the generated power of the cell stack 6 is kept at the maximum output (at this time, a state with little temperature fluctuation is maintained over time), and such a stable state (for example, The detected temperature of the temperature detecting means 56 is kept within a predetermined temperature range) for a certain time (for example, 2 to 5 hours) or the detected current of the current detecting means 58 is within the predetermined current width. When a condition such as whether a certain time (for example, 2 to 5 hours) has passed is maintained, the deterioration determination mode switching means 82 switches to the operation in the deterioration determination mode (step S22).

  When the operation is switched to the deterioration determination mode, the first and second voltage detection means 52 and 54 detect the voltage between the first and second parts of the cell stack 6 (step S23). Then, insertion control is performed to insert the output current of the cell stack 6 to the lower side (step S24). In this embodiment, the output current insertion control is performed over five stages. In the first stage, the output current is pulled up to 80%, in the second stage, the output current is pulled up to 60%, and in the third stage. The output current is controlled to be pulled up to 45%, the output current is pulled up to 35% in the fourth stage, and the output current is pulled up to 25% in the fifth stage. When the first-stage insertion / extraction control is performed, the process proceeds from step S24 to step S25, and the first and second voltage detection means 52 and 54 detect the voltage between the first and second parts of the cell stack 6, This detected voltage is stored in the memory means 70A as the voltage at the time of the first stage insertion. When the first stage of insertion is performed, the process returns from step S26 to step S24, then the second stage of insertion control is performed, and the first and second voltage detection means 52, 54 are connected to the first of the cell stack 6. And the voltage between the second parts is detected again, and this detected voltage is stored in the memory means 70A as the voltage at the time of the second stage insertion. Such insertion control is performed up to the fifth stage, and the detected voltages of the first and second voltage detecting means 52 and 54 up to the fifth stage are stored in the memory means 70A.

  When the insertion / subtraction control up to the fifth stage is performed in this way, the process proceeds from step S26 to step S27, and based on the detected current and the detected voltage stored in the memory unit 70A, the virtual open circuit voltage calculating unit 84 performs the virtual control. An open circuit voltage calculation is performed. For the first part of the cell stack 6, the detection current of the current detection means 58 and the detection voltage of the first voltage detection means 52 in the operation in the normal mode, the detection current and the detection voltage at the first stage of pulling, Plot of current-voltage characteristics (IV characteristics) in the first part of the cell stack 6 based on the detected current and the detected voltage in the second stage (and the third stage, the fourth stage, and the fifth stage). As a linear function, and the first virtual open circuit voltage (that is, the first virtual intercept) is calculated using the obtained linear function. Further, for the second part of the cell stack 6, the detection current of the current detection means 58 and the detection voltage of the second voltage detection means 54 in the normal mode operation, the detection current and the detection at the time of the first stage insertion. Based on the voltage and the detected current and the detected voltage in the second stage (and the third stage, the fourth stage, and the fifth stage), the current-voltage characteristic (IV characteristic) in the second part of the cell stack 6 Is obtained as a linear function, and the second virtual open circuit voltage (that is, the second virtual intercept) is calculated using the obtained linear function.

Also in this embodiment , as in the above-described reference embodiment , one of the fuel cells 8 obtained by dividing the detection voltages of the first and second voltage detection means 52 and 54 by the number of fuel cells 8 in those portions. Per unit detection voltage (unit detection voltage) is calculated, and the first virtual unit open circuit voltage per fuel cell 8 (that is, the first virtual unit intercept) is calculated using the unit detection voltages of the first to fifth stages. ) And the second virtual unit open circuit voltage (that is, the second virtual unit intercept) are calculated, and the first and second virtual unit open circuit voltages are preferably used as the first and second virtual open circuit voltages.

  In this embodiment, when the control of the output current of the cell stack 6 is performed, it is performed over five steps on the lowering side. However, it may be performed on the lowering side by one to four steps, or on the lowering side by six or more steps. Further, the output current to be inserted need not be up to 25% of the output current, and may be controlled to be about 50%, for example. In addition, the time required for the insertion control may be about 10 to 30 seconds, for example, and the insertion control can be performed by limiting the output current of the inverter 41, for example.

  When the first and second virtual open circuit voltages for the first and second portions of the cell stack 6 are calculated as described above, the process proceeds to step S28, where the voltage difference calculation means 64 is connected to the first and second virtual open circuit voltages. If the voltage difference of the circuit voltage is calculated and this voltage difference is smaller than the predetermined voltage value, the process proceeds from step S29 to step S30, and the fuel cell system continues to operate in the normal mode, assuming that local degradation of the cell stack 6 has not occurred. In the same manner as described above, the operation is controlled so that the fuel utilization rate in the cell stack 6 is about 70 to 75%.

  Further, if the voltage difference calculated by the voltage difference calculation means 64 (that is, the voltage difference between the first and second virtual open circuit voltages) is equal to or greater than a predetermined voltage value, it is assumed that local degradation has occurred in the cell stack 6. Shifting from S29 to step S31, the deterioration determining means 66 determines that the cell stack 6 is in a deteriorated state, and generates a deterioration signal. Thus, the mode switching means 68 generates a mode switching signal based on this deterioration signal, sets the deterioration mode based on this mode switching signal, and the fuel cell system is operated in the deterioration mode, as described above. The operation is controlled so that the fuel utilization rate in the cell stack 6 is lower than that in the normal mode, for example, about 65 to 70%.

  The solid oxide fuel cell system shown in FIGS. 6 and 7 is used to switch to the deterioration determination mode during power generation with the maximum output of the fuel cell system, and the first and second voltage detection means 52 and 54 during the pulling control. A virtual open circuit voltage was calculated based on the detected voltage, and an endurance test was performed to confirm whether the deterioration state could be diagnosed using the voltage difference between the virtual open circuit voltages. Even in this test, the detection voltage per fuel cell (that is, the detection voltage of the first and second voltage detection means 52, 54 divided by the number of fuel cells 8 in the first and second parts) (that is, Unit detection voltage), the first and second virtual unit open circuit voltages are calculated based on the unit detection voltage, and the voltage difference between the first and second virtual unit open circuit voltages is calculated.

  FIG. 8 shows a change in the detected voltage between the first and second portions when the deterioration determination mode is switched to during the operation of the fuel cell system and the output current of the cell stack 6 is controlled to be pulled down. . In this experiment, the output current of the cell stack 6 at the rated (maximum output) was set to 100%, and the insertion control was performed in five stages of 80%, 60%, 45%, 35%, and 25%. . This insertion control was performed over 10 seconds, and the voltage between the first and second parts during the insertion control was detected by the first and second voltage detection means 52 and 54.

  The relationship (IV characteristic) between the detected voltage between the first and second parts of the cell stack 6 and the output current of the cell stack 6 when the insertion / extraction control is performed is as shown in FIG. Based on the IV characteristics obtained for the second part and the second part, the first and second virtual open circuit voltages can be calculated as follows.

  That is, the IV characteristics for the first and second parts can be obtained as a linear function as understood from FIG. 8, and the first and second virtual intercepts (ie, , First and second virtual open circuit voltages) can be calculated.

For example, when the output current of the cell stack 6 is measured at two points of 100% and X% (X = 0 to 50%) and the detected voltages are V100 and VX, the virtual open circuit voltage V is
V = [V100 + (V100−VX) X100] / (100−X)
Can be calculated as

  The pulling control for pulling the output current to the lower side need not be performed only when the temperature of the cell stack 6 or the output current is in a stable state as described above. For example, an appropriate time of 2 hours or more Periodic insertion / extraction control may be performed at intervals to diagnose the deterioration state of the cell stack 6, or only when the temperature of the cell stack 6 and the output current are within a predetermined range, intercept calculation is performed. The calculation may be performed and stored.

The first and second virtual open circuit voltages (first and second virtual unit open circuit voltages) and the voltage difference between those operated for about 18,000 hours from the start of the test changed as shown in FIG. . The cell stack 6 mounted in this fuel cell system is the same as the cell stack 6 in the above-described reference embodiment , and it can be considered that the same deterioration has progressed. As understood from FIG. 9, the voltage difference (right axis in FIG. 9) of the first and second virtual open circuit voltages (first and second virtual unit open circuit voltages) is about −6 mV on average in the initial state. However, it can be seen that the decrease began to increase after 6500 hours. For this reason, it is possible to detect the deterioration state of the cell stack 6 even using the virtual open circuit voltage, and it is possible to detect the deterioration state earlier than when detecting the open circuit voltage at startup. It becomes possible.

  The reason is considered as follows. In the fuel electrode of the fuel battery cell 8 whose gas sealing performance has deteriorated, it is considered that a sufficient amount of reducing gas cannot be supplied in the power generation state, and a partial oxidation state occurs. In the fuel battery cell 8 in this state, It is considered that an electromotive force slightly smaller than the original electromotive force is generated in the process of reducing the current to a low current. On the other hand, at the time of start-up, since the fuel is sufficiently supplied to the fuel electrode, even if an oxidation state is generated in the fuel electrode in the fuel cell 8 having a deteriorated gas sealing performance, the fuel electrode 8 is reduced again in the process of temperature rise. It is considered that the original electromotive force has been restored. From these facts, it is presumed that the abnormality detection with high sensitivity can be performed by performing the pulling operation from the power generation state to the low current side.

  The embodiment of the solid oxide fuel cell system according to the present invention has been described above. However, the present invention is not limited to such an embodiment, and various changes or modifications can be made without departing from the scope of the present invention. It is.

For example, in the above-described embodiment , the voltage difference of the virtual open circuit voltage between the first and second portions of the cell stack is used, but instead of this voltage difference, the voltage ratio thereof may be used. Even in this case, the same effect as described above can be achieved.

2 Solid oxide fuel cell system 4 Reformer 6 Cell stack 8 Fuel cell 52 First temperature detecting means 54 Second temperature detecting means 56 Temperature detecting means 58 Current detecting means 60, 60A Control means 64 Voltage difference calculating means 66 Degradation determination means 68 Mode switching means 82 Deterioration determination mode switching means 84 Virtual open circuit voltage calculation means

Claims (3)

A cell comprising a reformer for reforming raw fuel gas, and a plurality of fuel cells for generating power by oxidation and reduction of the reformed fuel gas and the oxidizing material reformed by the reformer A stack, a fuel pump for supplying raw fuel gas to the reformer, an oxidant blower for supplying an oxidant to the cell stack, and a control for controlling the fuel pump and the oxidant blower A solid oxide fuel cell system comprising:
A first voltage detecting means for detecting a voltage between the first parts of the cell stack; and a second voltage detecting means for detecting a voltage between the second parts of the cell stack, the control The means includes a deterioration determination mode switching means for switching to a deterioration determination mode for determining a deterioration state of the cell stack, a virtual open circuit voltage calculating means for calculating a virtual open circuit voltage, and a deterioration state of the cell stack. Including a deterioration determining means for determining,
When switched to the deterioration determination mode by the deterioration determination mode switching means, the control means performs insertion control to the side where the output current of the cell stack becomes lower, and the first voltage detection means An output voltage between the first parts of the cell stack is detected, and the second voltage detection means detects an output voltage between the second parts of the cell stack during insertion control, and the virtual open circuit The circuit voltage calculation means calculates a first virtual open circuit voltage based on the detection voltage of the first voltage detection means, and calculates a second virtual open circuit voltage based on the detection voltage of the second voltage detection means. The deterioration determination means determines the deterioration state of the cell stack based on the first and second virtual open circuit voltages. Voltage detection means is provided corresponding to each of the three or more parts of the cell stack, each voltage detection means detects a voltage between the corresponding parts, and the deterioration determination means of the control means is the voltage detection 2. The solid oxide fuel cell system according to claim 1, wherein a deterioration state of the cell stack is determined using a detection voltage of the means. The control means further includes a mode switching means for switching the operation mode. When the deterioration determination means determines that the operation is in a deteriorated state, the mode switching means switches to the operation in the deterioration mode, and in the operation in the deterioration mode. , the control unit solid oxide fuel cell system according to claim 1 or 2, characterized in that controls the operation of the system so that the fuel utilization in the cell stack is lowered.

Images