Catalyst degradation suppression equipment

Abstract

A catalyst degradation suppression device, comprising: a first device for obtaining a voltage V of a fuel cell (= a catalyst voltage V cat ) as a variable for estimating a response speed (time constant τ), at which the coverage of the oxide film of the catalyst particles contained in the cathode of the fuel cell changes; a second device for reading out the time constant τ corresponding to the voltage V at the current time t from the pre-made map A t , the pre-made map A represents the relationship between the voltage V and the time constant τ and corresponds to the catalyst particles; a third device for using the time constant τ t Generate a continuous time dynamic filter F(s,τ), and convert the continuous time dynamic filter F(s,τ) into a discrete time dynamic filter F(z,τ); and a fourth device for inputting a target voltage V to the discrete time dynamic filter F(z,τ) r And output the correction target voltage V r‑fil .

Description

Technical Field

[0001] The present invention relates to a catalyst degradation suppression device, and more particularly, to a catalyst degradation suppression device for suppressing degradation of catalyst particles on a cathode side in a transient state where voltage changes rapidly. Background Art

[0002] A polymer electrolyte fuel cell has a membrane electrode assembly (MEA) in which electrodes including a catalyst are bonded to both surfaces of an electrolyte membrane. Each electrode generally has a double-layer structure including a catalyst layer and a gas diffusion layer including a catalyst. The gas diffusion layer is used to supply a reaction gas and electrons to the catalyst layer, and carbon paper, carbon cloth, etc. are used. In addition, the catalyst layer is a portion that acts as a reaction field for an electrode reaction, and generally includes a composite of carbon carrying catalyst particles such as platinum and a solid polymer electrolyte (catalyst layer ionomer).

[0003] Current collectors (also referred to as separators) having air flow paths are further placed on both surfaces of the MEA. A polymer electrolyte fuel cell generally has a structure (fuel cell stack) in which a plurality of unit cells including the MEA and current collectors are stacked.

[0004] In polymer electrolyte fuel cells, the dissolution rate of catalyst particles on the cathode side increases as the voltage increases. Conversely, when the voltage increases, the surface of the catalyst particles on the cathode side is covered with an oxide film, thereby inhibiting the dissolution of the catalyst particles. However, when the voltage increases rapidly, the oxide film is delayed in formation, so when rapid voltage fluctuations are repeated, the elution of the catalyst particles gradually develops.

[0005] The eluted catalyst components are precipitated as particles in the electrolyte membrane, or precipitated on the surface of other catalyst particles present in the catalyst layer, and the catalyst particles are coarsened. One or both of the elution of the catalyst components outside the catalyst layer and / or the coarsening of the catalyst particles lead to degradation of fuel cell performance. In addition, such degradation of the catalyst can also occur in fuel cells other than polymer electrolyte fuel cells.

[0006] To solve this problem, various proposals have been made so far.

[0007] For example, Patent Document 1 discloses a fuel cell operating method in which, when a target voltage of the fuel cell is higher than a voltage before fluctuation and is equal to or higher than a dissolution start voltage at which a catalyst may dissolve,

[0008] (a) increasing the voltage to a first level rising voltage, which is a voltage at which the catalyst is covered with a protective film or higher, and

[0009] (b) After a predetermined time has elapsed since the voltage reached the first-step rising voltage, the voltage is increased to the target voltage.

[0010] The document describes that when the catalyst may be dissolved, by controlling the operation of the fuel cell so that the voltage can reach the target voltage after waiting for the formation of a protective film on the catalyst, the catalyst metal can be prevented from deteriorating.

[0011] When the method described in Patent Document 1 is used, the elution of catalyst particles and the degradation of fuel cell performance due to the elution can be suppressed to a certain extent. However, in actual situations, the power demand from the driver varies, and thus the power of the fuel cell also varies differently. Therefore, by the method of raising the voltage to the target voltage only after a predetermined time has elapsed as described in Patent Document 1, it is difficult to make them compatible to prevent catalyst degradation and meet the differently varying power demand from the driver.

[0012] Furthermore, the degradation rate of catalyst particles varies not only according to the voltage of the fuel cell but also according to the type of catalyst particles, the degradation state of the catalyst particles, the use environment (temperature, humidity) of the catalyst particles, etc. However, a method of suppressing the degradation of catalyst particles in consideration of these points has never been proposed.

[0013] [Citation List]

[0014] [Patent Document]

[0015] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2010-067434 Summary of the invention

[0016] Problems to be solved by the present invention

[0017] A problem to be solved by the present invention is to provide a catalyst degradation suppression device capable of suppressing the degradation of catalyst particles on the cathode side in a transient state where the voltage changes rapidly.

[0018] Another problem to be solved by the present invention is to provide a catalyst degradation suppression device capable of suppressing degradation of catalyst particles on the cathode side even when the type of catalyst particles, the degradation state of catalyst particles, or the use environment of catalyst particles is different.

[0019] Technical means of solving problems

[0020] In order to solve the above problems, a catalyst degradation suppression device according to the present invention includes:

[0021] The first device is used to obtain the voltage V of the fuel cell (=catalyst voltage V cat) as a variable for estimating a response speed (time constant τ) at which the coverage of an oxide film of catalyst particles contained in a cathode of a fuel cell changes;

[0022] The second device is used to read the time constant τ corresponding to the voltage V at the current time t from the pre-made map A. t , the prefabricated map A represents the relationship between the voltage V and the time constant τ and corresponds to the catalyst particles;

[0023] The third means is used to t Generate a continuous-time dynamic filter F(s,τ), and convert the continuous-time dynamic filter F(s,τ) into a discrete-time dynamic filter F(z,τ); and

[0024] The fourth device is used to input the target voltage V to the discrete time dynamic filter F(z,τ) r And output the correction target voltage V r-fil .

[0025] Effects of the Invention

[0026] In a fuel cell, the dissolution rate of the catalyst particles on the cathode side increases with increasing voltage. On the contrary, when the voltage increases, the surface of the catalyst particles on the cathode side is covered with an oxide film, thereby suppressing the dissolution of the catalyst particles. However, when the voltage increases rapidly, the oxide film is delayed in formation, so the elution of the catalyst particles develops. In this case, the response speed of the change in the coverage of the oxide film of the catalyst particles (i.e., the dissolution rate of the catalyst particles) mainly depends on the size of the voltage V of the fuel cell and the size of the rate of change of the voltage V.

[0027] On the other hand, the elution of catalyst particles due to delayed formation of the oxide film can be suppressed by:

[0028] (a) Read the time constant τ corresponding to the voltage V at the current time t from the pre-made map A t , the pre-made map A represents the relationship between the voltage V and the response speed (time constant τ) and corresponds to the catalyst particles used in the fuel cell;

[0029] (b) Using τ t Generate a continuous-time dynamic filter F(s,τ); and

[0030] (c) The voltage V is corrected by using F(s, τ) (in other words, the actual voltage change rate is made slower than the required voltage change rate).

[0031] Furthermore, the time constant τ depends not only on the voltage V but also on the voltage decrement ΔV (i.e., the degree of degradation of the catalyst particles), the temperature T of the fuel cell, and the humidity RH of the gas supplied to the cathode and the anode, respectively. Therefore, by preparing corresponding maps corresponding to these variables in advance, reading out the correction gain corresponding to each of these variables from the map, and correcting the time constant τ using the read correction gain, the elution of the catalyst particles can be further suppressed. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a schematic diagram showing the change of the degradation index m1 of the catalyst particles over time when the voltage V is changed stepwise.

[0033] Figure 2 is a schematic diagram showing the change of dissolution rate F(V) and coverage θ over time when the voltage V is changed stepwise.

[0034] Figure 3 is a schematic diagram of the voltage dependence of the transfer characteristics of the catalyst particles.

[0035] Figure 4 is a schematic diagram of the transfer characteristics of simple modeling.

[0036] Figure 5 It is a schematic diagram of the model of the transfer characteristic (control object).

[0037] Figure 6 is a schematic diagram of a method for suppressing an increase in a degradation index.

[0038] Figure 7 It is a schematic diagram of a method for measuring the response speed (time constant τ) by bench experiment.

[0039] Figure 8 is a schematic diagram of a method for installing a catalyst degradation suppressing device according to the present invention.

[0040] Fig. 9 is a schematic diagram of a method for handling multiple voltages.

[0041] Fig.10 is a schematic diagram of changes in transfer characteristics when the catalyst deteriorates.

[0042] Fig.11 is a schematic diagram of a method for treating catalyst degradation.

[0043] Fig.12 is a schematic diagram of a method for handling temperature changes and / or humidity changes.

[0044] Fig.13 is a schematic diagram of the voltage change during the transient time. DETAILED DESCRIPTION

[0045] The embodiments according to the present invention are described in detail below.

[0046] [1. Catalyst degradation suppression equipment]

[0047] The catalyst degradation suppression device according to the present invention comprises:

[0048] The first device is used to obtain the voltage V of the fuel cell (=catalyst voltage V cat ) as a variable for estimating a response speed (time constant τ) at which the coverage of an oxide film of catalyst particles contained in a cathode of a fuel cell changes;

[0049] The second device is used to read the time constant τ corresponding to the voltage V at the current time t from the pre-made map A. t , the prefabricated map A represents the relationship between the voltage V and the time constant τ and corresponds to the catalyst particles;

[0050] The third means is used to t Generate a continuous-time dynamic filter F(s,τ), and convert the continuous-time dynamic filter F(s,τ) into a discrete-time dynamic filter F(z,τ); and

[0051] The fourth device is used to input the target voltage V to the discrete time dynamic filter F(z,τ) r And output the correction target voltage V r-fil .

[0052] [1.1. Application Target]

[0053] The catalyst degradation suppression device according to the present invention is applicable to any fuel cell that may cause catalyst degradation when voltage fluctuation occurs. Fuel cells to which the present invention is applied are, for example, polymer electrolyte fuel cells, phosphate fuel cells, solid oxide fuel cells, molten carbonate fuel cells, and alkaline fuel cells.

[0054] The components of the catalyst particles are not particularly limited, and the optimal components can be selected according to the purpose. In the case of a polymer electrolyte fuel cell, examples of the catalyst particles are:

[0055] (a) Noble metals (Pt, Au, Ag, Pd, Rh, Ir, Ru and Os);

[0056] (b) alloys containing two or more noble metal elements; and

[0057] (c) Alloys containing one or more noble metal elements and one or more base metal elements (e.g., Fe, Co, Ni, Cr, V and Ti).

[0058] [1.2. First device]

[0059] The first device comprises a device for obtaining a voltage V of the fuel cell (=catalyst voltage V cat ) is used as a variable for estimating a response speed (time constant τ) at which the coverage of the oxide film of the catalyst particles contained in the cathode of the fuel cell changes.

[0060] The first device may also include the following as a variable for estimating the response speed (time constant τ):

[0061] (a) a device for obtaining a voltage reduction ΔV;

[0062] (b) means for obtaining the temperature T of the fuel cell; and / or

[0063] (c) Means for obtaining the humidity RH of the gases supplied to the cathode and anode, respectively.

[0064] [1.2.1. Obtain voltage V]

[0065] The dissolution rate of catalyst particles changes mainly depending on the magnitude of the voltage V in the fuel cell and the magnitude of the rate of change of the voltage V. Therefore, in order to suppress degradation of catalyst particles, the first device must have at least means for obtaining the voltage V in the fuel cell as a variable for estimating τ.

[0066] A fuel cell generally includes a unit cell having a membrane electrode assembly (MEA) in which electrodes are bonded on both surfaces of an electrolyte membrane, and current collectors arranged on both surfaces of the MEA.

[0067] In the present invention, "voltage V" refers to the catalyst voltage V cat , unless otherwise specified.

[0068] "Catalyst voltage V cat " refers to the potential difference between the reference electrode installed on the cathode side catalyst layer and the cathode side current collector in the unit cell. "Inter-terminal voltage V cell ” refers to the potential difference between the cathode side collector and the anode side collector in the unit cell. When the current flowing in the unit cell is defined as I and the resistance of the unit cell is defined as R, the relational expression V(V cat )=V cell +IR.

[0069] It is preferred to use the actual measured V cat As the voltage V used to estimate τ. This is because τ is related to V cat When actually measuring V cell When V cell To estimate Vcat , and estimation errors may occur in this case, as will be described below.

[0070] For actually measuring or estimating the voltage V (i.e., the catalyst voltage V cat Examples of devices that are:

[0071] (a) Device A, for obtaining the terminal voltage V of the fuel cell cell , current I and resistance R, and according to the relational expression V = V cell +IR to calculate the voltage V;

[0072] (b) Device B, for obtaining V cell , V cell Substitute the previously obtained V cell The relationship between the voltage V and the voltage V = f(V cell ), and estimate the voltage V; and

[0073] (c) A device C for installing a reference electrode in advance on the cathode-side catalyst layer of the fuel cell and actually measuring the voltage V by using the potential difference between the reference electrode and the cathode-side current collector.

[0074] In the case of device A, in addition to the inter-terminal voltage V cell In addition, it is also necessary to obtain the current I and the resistance R. Generally, since the current collector is exposed to the outer surface of the fuel cell stack, it is relatively easy to obtain V cell When V cell When I is , I can be obtained. R can be obtained by impedance measurement. Device A is easy and convenient, and because it uses V which is relatively easy to measure cell , I and R are used to estimate the voltage V, which has the advantage of relatively high estimation accuracy.

[0075] In the case of device B, only the inter-terminal voltage V is obtained cell , and according to the previously obtained V cell The voltage V is estimated by using the relational expression between θ and the voltage V. In terms of the estimation accuracy of the voltage V, the device B is not as good as the device A, but compared with the device A, the device B estimates the voltage V more simply and conveniently.

[0076] In the case of device C, the voltage V is actually measured by using a reference electrode. Compared with device A, device C can obtain a more accurate voltage V, but has a disadvantage in that the structure of the fuel cell is complicated.

[0077] [1.2.2. Obtaining voltage reduction △V]

[0078] The first device may also include a device for obtaining a voltage reduction ΔV (ie, a catalyst voltage V cat The voltage reduction △Vcat ) as a means of estimating the variable τ.

[0079] Here, “voltage reduction ΔV” refers to the difference between the initial state voltage V0 and the voltage V after t hours of use at the reference current I0. t The difference between (=V t -V0). The reference current I0 is not particularly limited, and an optimal value can be selected according to the purpose.

[0080] The power demand on the fuel cell is high, so during normal operation the current I sometimes momentarily becomes the reference current I0. In this case, the change of ΔV with time can be obtained by sequentially storing the voltage V.

[0081] △V is an index indicating the degree of degradation of catalyst particles. If sudden voltage fluctuations are repeated during the operation of the fuel cell, the following phenomenon occurs: catalyst particles with smaller particle sizes are preferentially eluted and precipitated on the surface of catalytic particles with larger particle sizes (i.e., a phenomenon of catalyst particle coarsening). If this phenomenon occurs, the voltage V of the reference current I0 decreases. As a result, τ changes even if the voltage V does not change. If τ is therefore estimated by using only the voltage V, then in some cases, the estimation accuracy of τ may be reduced. On the contrary, by obtaining a voltage decrement △V in addition to the voltage V and correcting τ based on △V, the degradation of the catalyst particles can be further suppressed.

[0082] [1.2.3. Get temperature T]

[0083] The first device may further comprise means for obtaining the temperature T of the fuel cell as a variable for estimating τ.

[0084] The temperature of the fuel cell varies according to the use environment and power demand of the fuel cell. Meanwhile, each of the elution reaction of the catalyst particles and the formation reaction of the oxide film is a chemical reaction, so the reaction rate depends on the temperature T. Therefore, if τ is estimated only by the voltage V, the estimation accuracy of τ may sometimes decrease. On the contrary, by obtaining the temperature T in addition to the voltage V and correcting τ according to T, the degradation of the catalyst particles can be further suppressed.

[0085] [1.2.4. Get humidity RH]

[0086] The first device may further include means for obtaining humidity RH of gases respectively supplied to the cathode and the anode as a variable for estimating τ.

[0087] For example, Pt particles are used as cathode catalysts in polymer electrolyte fuel cells, but when used, the Pt particles sometimes coarsen. Water is needed to advance the reaction, so the reaction rate depends on the humidity RH. Therefore, if τ is estimated only by the voltage V, the estimation accuracy of τ may sometimes decrease. On the contrary, by obtaining the humidity RH in addition to the voltage V and correcting τ according to RH, the degradation of the catalyst particles can be further suppressed.

[0088] [1.3. Second device]

[0089] The second device includes a device for reading a voltage Vτ (hereinafter also referred to as “voltage V t ”) corresponding to the time constant τ (hereinafter also referred to as “time constant τ t ”) device, the pre-made map A represents the relationship between the voltage V and the time constant τ and corresponds to the catalyst particles.

[0090] The second device may be a device further comprising:

[0091] (a) is used to read the voltage reduction ΔV (hereinafter also referred to as “voltage reduction ΔV”) at the current time t from the pre-made map B. t ”) corresponding to the correction gain k (hereinafter also referred to as “correction gain k t ”), and multiply the time constant τ by the correction gain k t The device, the pre-made map B represents the relationship between the voltage reduction ΔV and the correction gain k of the time constant τ and corresponds to the catalyst particles;

[0092] (b) is used to read the temperature T (hereinafter also referred to as "temperature T t ”) corresponding to the correction gain p (hereinafter also referred to as “correction gain p t ”), and multiply the time constant τ by the correction gain p t The prefabricated map C represents the relationship between the temperature T and the correction gain p of the time constant τ and corresponds to the catalyst particles; and / or

[0093] (c) is used to read the humidity RH (hereinafter also referred to as "humidity RH") corresponding to the current time t from the pre-made map D. t ”) corresponding to the correction gain r (hereinafter also referred to as “correction gain r t ”), and multiply the time constant τ by the correction gain r t The prefabricated map D represents the relationship between the humidity RH and the correction gain r of the time constant τ and corresponds to the catalyst particles.

[0094] [1.3.1. Reading the time constant τ t ]

[0095] When the degradation degree and use conditions of the catalyst particles are the same, the time constant τ at a specific voltage V depends on the composition of the catalyst particles. Therefore, a map A showing the relationship between V and τ is prepared in advance for each composition of the catalyst particles and stored in a memory.

[0096] In the memory, only a map A corresponding to one catalyst particle may be stored, or a plurality of maps A corresponding to two or more catalyst particles may be stored. When a plurality of maps A are stored in the memory, a specific map A corresponding to the catalyst particles used in the fuel cell is selected from the plurality of maps A.

[0097] Subsequently, when V is obtained by the first device t When V t The corresponding τ t When there is no V t The corresponding τ t When, you can use the V t Multiple time constants τ corresponding to multiple voltages V in the vicinity t To apply linear interpolation. t Used to generate a dynamic filter F(s,τ) in a third device to be described below.

[0098] [1.3.2. Correction of τ by voltage reduction △V]

[0099] The reaction rate of the degradation reaction of the catalyst particles also depends on the voltage reduction ΔV (the degree of degradation of the catalyst particles), and the time constant τ also changes in response thereto. If the degree of change of τ is defined as the correction gain k in this case, k also depends on the components of the catalyst particles. Therefore, it is desirable to prepare in advance a map B indicating the relationship between ΔV and k for each component of the catalyst particles and store it in a memory.

[0100] In the memory, only a map B corresponding to one catalyst particle may be stored, or a plurality of maps B corresponding to two or more catalyst particles may be stored. When a plurality of maps B are stored in the memory, a specific map B corresponding to the catalyst particles used in the fuel cell is selected from the plurality of maps B.

[0101] Then, when △V is obtained by the first device t When K is read from the map B t When there is no t The corresponding K t When thet Multiple correction gains K corresponding to multiple voltage reductions △V in the vicinity t To apply linear interpolation. When reading K t When τ (including the p t and / or t Corrected τ) multiplied by K t . Will pass k t Corrected τ(=k t ×τ) is used to generate a dynamic filter F(s,τ) in a third device to be described below.

[0102] [1.3.3. Correction of τ by temperature T]

[0103] The reaction rate of the degradation reaction of the catalyst particles also depends on the temperature T, and the time constant τ also changes in response thereto. If the degree of change of τ is defined as the correction gain p in this case, p also depends on the components of the catalyst particles. Therefore, it is desirable to prepare in advance a map C indicating the relationship between T and p for each component of the catalyst particles and store it in a memory.

[0104] In the memory, only a map C corresponding to one catalyst particle may be stored, or a plurality of maps C corresponding to two or more catalyst particles may be stored. When a plurality of maps B are stored in the memory, a specific map C corresponding to the catalyst particles used in the fuel cell is selected from the plurality of maps C.

[0105] Subsequently, when T is obtained by the first device t When T t The corresponding p t When there is no T in the mapping graph C t The corresponding p t When using t Multiple correction gains p corresponding to multiple temperatures T in the vicinity t To apply linear interpolation. When reading p t When τ (including k described below) is t and / or t Corrected τ) multiplied by p t . Will be through p t Corrected τ(=p t ×τ) is used to generate a dynamic filter F(s,τ) in a third device to be described below.

[0106] [1.3.4. Correction of τ by RH]

[0107] The reaction rate of the degradation reaction of the catalyst particles also depends on the humidity RH, and the time constant τ also changes in response thereto. If the degree of change of τ is defined as the correction gain r in this case, r also depends on the components of the catalyst particles. Therefore, it is desirable to prepare in advance a map D representing the relationship between RH and r for each component of the catalyst particles and store it in a memory.

[0108] In the memory, only a map D corresponding to one catalyst particle may be stored, or a plurality of maps D corresponding to two or more catalyst particles may be stored. When a plurality of maps B are stored in the memory, a specific map D corresponding to the catalyst particles used in the fuel cell is selected from the plurality of maps D.

[0109] Then, when RH is obtained by the first device t When RH is read from the map D t The corresponding r t When there is no RH in the mapping diagram D t The corresponding r t When using the same t Multiple correction gains r corresponding to multiple humidity RHs in the vicinity t To apply linear interpolation. When reading r t When τ (including the k t and / or t Corrected τ) multiplied by r t . Will be through r t Corrected τ(=r t ×τ) is used to generate a dynamic filter F(s,τ) in a third device to be described below.

[0110] [1.4. Third device]

[0111] The third means comprises a device for using the time constant τ t A device for generating a continuous-time dynamic filter F(s,τ) and converting the continuous-time dynamic filter F(s,τ) into a discrete-time dynamic filter F(z,τ).

[0112] The third device may be a device further comprising:

[0113] (a) For the correction gain k t A device for generating a continuous-time dynamic filter F(s,τ) by correcting the time constant τ and converting the continuous-time dynamic filter F(s,τ) into a discrete-time dynamic filter F(z,τ);

[0114] (b) For the correction gain p tA device for generating a continuous-time dynamic filter F(s,τ) by correcting the time constant τ and converting the continuous-time dynamic filter F(s,τ) into a discrete-time dynamic filter F(z,τ); and / or

[0115] (c) is used by using the correction gain r t A device for generating a continuous-time dynamic filter F(s, τ) by correcting the time constant τ and converting the continuous-time dynamic filter F(s, τ) into a discrete-time dynamic filter F(z, τ).

[0116] When the voltage V changes suddenly, the dissolution rate of the catalyst particles increases instantaneously (overshoots). The dynamic filter F(s,τ) must be able to be used by using τ (including the correction gain k t 、p t and r t The dynamic filter F(s,τ) is not particularly limited as long as it exhibits such a function.

[0117] An example of such a dynamic filter F(s, τ) is represented by the following expression (4) or (6). The details of expressions (4) and (6) will be described below.

[0118] F(s,τ)=(τ·s+1) / (2τ·s+1) (4)

[0119] Here, s is the Laplace operator.

[0120] F(s,τ)=G m (s) / G(s,τ) (6)

[0121] Here, G m (s) is the transfer function of the ideal response model, and G(s,τ) is the transfer function of the model.

[0122] In addition, the continuous time type dynamic filter F(s, τ) is converted into a discrete time type dynamic filter F(z, τ). The method of converting F(s, τ) into F(z, τ) is not particularly limited and a known method can be used. An example of a method of converting F(s, τ) into F(z, τ) will be described below.

[0123] [1.5. The fourth device]

[0124] The fourth device includes a device for inputting a target voltage V to F(z,τ) r And output the correction target voltage V r-fil The target voltage V r-fil is sent to the actuator controller, and the control object is controlled by the controller.

[0125] [2. Example]

[0126] An example of the catalyst degradation suppressing device according to the present invention is described below.

[0127] [2.1. Target catalyst degradation]

[0128] Figure 1 Schematic diagram showing the change of the degradation index m1 of the catalyst particles with time when the voltage V changes stepwise. The catalyst particles have Figure 1 Degradation characteristics shown. Figure 1 The upper row shows the voltage V of the fuel cell (i.e., the catalyst voltage V cat ) Example of gradual increase. Figure 1 The lower row shows how the degradation index m1 indicating the degree of degradation of the catalyst particles changes with time due to the voltage V input to the upper row.

[0129] The "degradation index m1" refers to the degree to which the performance of catalyst particles (eg, platinum particles) used in a fuel cell deteriorates over time. In the present invention, the degradation index is the dissolution rate of the catalyst particles. The dissolution rate is represented by the following expression (1).

[0130] m1=F(V)×(1-θ(V,t)) (1)

[0131] here,

[0132] m1 is the degradation index (dissolution rate),

[0133] F(V) is the dissolution rate of the catalyst depending on the voltage V,

[0134] θ is the coverage (the ratio of the area of ​​the oxide film covering the surface of the catalyst particle to the surface area of ​​the catalyst particle, θ = 0 to 1) and

[0135] t is the time.

[0136] Figure 2 Schematic diagram showing the change of dissolution rate F(V) and coverage θ over time when voltage V is changed stepwise. Figure 2 is based on Figure 1 A view of F(V) and θ on the right side of expression (1) is plotted. Figure 2 As shown in the upper row, the dissolution rate F(V) of the catalyst increases with the increase of voltage V. At the same time, Figure 2 As shown in the lower row, when the voltage V increases, the oxide film is delayed in forming on the surface of the catalyst particles. When the oxide film is formed, as shown in expression (1), the dissolution of the catalyst particles, that is, the degradation, is suppressed. Obviously, Figure 1 The waveform in the lower row can be based on Figure 2 And expression (1) is obtained.

[0137] From the viewpoint of efficiency, it is necessary to set the voltage V to a target value V with high response according to the operating conditions. r At the same time, when the voltage V becomes high response (stepwise change), there occurs a problem that the degradation index (dissolution rate) temporarily increases and the degradation progresses, such as Figure 1 Therefore, in the present invention, a dynamic filter is used to suppress Figure 1 The instantaneous increment of the degradation index is shown in the shaded area in the lower row.

[0138] [2.2. Simple modeling of transfer characteristics]

[0139] Figure 3 Schematic diagram showing the voltage dependence of the transfer characteristics of the catalyst particles. F(V) and θ(V,t) on the right side of expression (1) are nonlinear functions (see, for example, reference 1). It has been found that depending on the value of the voltage V (in Figure 1 The transfer characteristics V→m1(V→F, V→θ) of the catalyst particles are determined by the value after time t1 in the equation ( ) and the material properties of the catalyst, as Figure 3 This means that when determining the voltage V and the material properties of the catalyst particles, Figure 3 As shown in the lower row, the shape of the time series waveform (response pattern) is determined.

[0140] [Reference 1] M. Darling and Jeremy P. Meyers, "Kinetic Model of Platinum Dissolution in PEFCs", Journal of the Electrochemical Society, Robert 150(11), A1523-A1525 (2003)

[0141] In the present invention, by using the above-mentioned knowledge, modeling of transfer characteristics is performed through the following steps.

[0142] Step 1: Determine the material properties of the catalyst.

[0143] Step 2: Determine the voltage V (with Figure 1 For example, if the lowest voltage is 0.60 V and the highest voltage is 0.90 V, then the voltages V = 0.60, 0.65, 0.70, 0.75, 0.80, 0.85, 0.90 (V) are used.

[0144] Step 3: Obtain the time series waveforms of the transfer characteristics (V→F, V→θ) for each voltage determined in step 2, and model them. For example, the time series waveforms can be obtained by simulating the model shown in reference 1.

[0145] The following describes a method for modeling the transfer characteristics of step 3 above. Figure 4 A schematic diagram showing the transfer characteristics of simple modeling. In the present invention, a simple transfer function is used to fit Figure 1 The time series waveform of the degradation index m1 is shown in the lower row. Therefore, the transfer characteristics are considered as "transient characteristics" and "steady-state characteristics", as shown in Figure 4 shown.

[0146] That is, when the voltage V changes, the degradation index increases (overshoots) instantaneously (temporarily). The "transient characteristic" is approximated by a high-pass filter. In addition, Figure 1 The time series waveform of the degradation index m1 shown in the lower row also includes a “steady characteristic” in which the degradation index changes stepwise according to the voltage V. This is expressed in a form proportional to the voltage V.

[0147] As a result, when the voltage V is input, the degradation index m1 is modeled as “transient characteristics”+“steady-state characteristics.” When a simple modeling expression is used in this way, a dynamic filter can be easily obtained as will be described below.

[0148] Figure 5 Schematic diagram showing a model of a transfer characteristic (control object). Figure 4 The modeling expression of the transfer characteristic (= transient characteristic + steady-state characteristic) is represented by the following expression (2). The first term on the right side of expression (2) corresponds to the transient characteristic (high-pass filter), and the second term corresponds to the steady-state characteristic.

[0149] [Formula 1]

[0150]

[0151] here,

[0152] J is the actual response output (= degradation index m1),

[0153] G(s,τ) is the transfer function of the model,

[0154] s is the Laplace operator,

[0155] V is the voltage (or catalyst voltage),

[0156] K m is the steady-state gain of the degradation index with respect to voltage (steady-state characteristic), and

[0157] τ is the response speed (time constant) when the catalyst particles are covered with an oxide film.

[0158] [2.3. Derivation of continuous-time dynamic filter]

[0159] [2.3.1. Example 1]

[0160] Figure 6 A schematic diagram showing a method for suppressing an increase in a degradation index. Figure 6 A method for suppressing an increase in the degradation index in the model of the transfer characteristic (control object) of Expression (2) is described.

[0161] exist Figure 6 In the upper row, the value of the output of the actual response J (deterioration index m1) is calculated by inserting the dynamic filter F(s,τ) on the output side of the voltage V and inputting the output voltage value into the transfer characteristic model (expression (2)). The role of the dynamic filter is to suppress degradation by improving the value of J, in other words, by improving the time series waveform of the input voltage V so that the time series waveform of J can be expected (ideal waveform).

[0162] As an example, the ideal waveform should respond proportionally to the voltage V. The ideal response waveform should be J m ,like Figure 6 In this example, this means eliminating transient characteristics (overshoot), and if the transient characteristics are eliminated, the degradation index can be suppressed to a small value.

[0163] Figure 6 The actual response output J in the upper row and the ideal response output J in the lower row m Should be equal in order to eliminate the transient characteristics, therefore, this leads to the following expression (3). From expression (2) and expression (3), the dynamic filter F(s, τ) is represented by expression (4).

[0164] G(s,τ)·F(s,τ)=K m (3)

[0165] F(s,τ)=(τ·s+1) / (2τ·s+1) (4)

[0166] In fact, τ changes in response to the voltage V. When τ corresponding to the current time t is defined as “τ t ", the dynamic filter F(s,τ) at the current time t is expressed by the following expression (4').

[0167] F(s,τ)=(τ t ·s+1) / (2τ t ·s+1) (4')

[0168] The coefficient of the dynamic filter represented by expression (4) depends only on the response speed (time constant τ) of the cover. The time constant τ can be obtained not only from the simulation in reference 1 but also from bench tests.

[0169] Figure 7 A schematic diagram showing a method for measuring the response speed (time constant τ) by bench testing is shown. First, Figure 7 As shown in the upper row, the voltage V is gradually input in the non-power generation state. Then, as Figure 7 As shown in the lower row, the time series waveform of the current value when the voltage V is gradually input is measured and the response speed (time constant τ) of the coverage is obtained from the waveform.

[0170] [2.3.2. Example 2]

[0171] exist Figure 6 In the ideal response model, set it to "K m (proportional gain)”. However, the ideal response model is considered to be helpful for degradation suppression even though it is not “K m ”, as long as it has this spike-suppressing transfer property.

[0172] Specifically, when the transfer function of the ideal response model is defined as “G m (s)” and in expression (3) “G m (s)" instead of "K m ", the following expression (5) is obtained. Furthermore, expression (6) is derived from expression (5).

[0173] G(s,τ)·F(s,τ)=G m (s) (5)

[0174] F(s,τ)=G m (s) / G(s,τ) (6)

[0175] G m (s) is represented by, for example, the following expression (7.1) or (7.2).

[0176] G m (s) = K m / (τ m ·s+1) (7.1)

[0177] Here, τ m is the time constant of the filter (τ m is different from the time constant τ of the model).

[0178] G m (s) = K m ·ω m 2 / (s 2 +2ζ m ω m s+ω m 2 ) (7.2)

[0179] Here, m and ω m is the filter constant.

[0180] Expression (7.1) represents the main lag filter, and expression (7.2) represents the secondary lag filter. m Use as G m (s), the ideal response J m is equal to the actual response J, such as Figure 6 In contrast, when the primary lag filter or the secondary lag filter is used as G m (s), the theoretical response J m lags behind the actual response J. However, when using a primary lag filter or a secondary lag filter, catalyst degradation and other requirements (such as fuel consumption and battery degradation) can be weighed.

[0181] Generally, if catalyst degradation is suppressed, other requirements are degraded as a trade-off. Therefore, when designing, it is necessary to set which performance is emphasized. m (s) makes this possible. When J m The response waveform ratio Figure 6 When the response waveform is more gentle, the catalyst degradation is emphasized more. In this case, the catalyst degradation is suppressed, but on the other hand, the fuel consumption or the degree of battery degradation increases.

[0182] [2.4. Installation method on the controller]

[0183] In order to install the continuous time type dynamic filter F(s,τ) on the vehicle controller, the continuous time type dynamic filter F(s,τ) is changed to a discrete time type and converted into a digital filter (discrete time type dynamic filter F(z,τ)) by using Tustin transformation, zero-order hold, etc. Here, z is a delay operator.

[0184] Figure 8 A schematic diagram showing a method for installing a catalyst degradation suppression device according to the present invention is shown. Figure 8 As shown, the discrete time dynamic filter F(z,τ) is inserted into the target voltage V r Here, the target voltage V r Generated by the upstream controller (not shown in the figure). r After F(z,τ) is input, the corrected target voltage V is output r-fil . Correction target voltage V r-fil is sent to the actuator controller, and the control object is controlled by the controller.

[0185] The following describes a method for converting a continuous-time dynamic filter F(s, τ) to a discrete-time dynamic filter F(z, τ). There are several conversion methods. Here, Tustin transformation of expression (8) is described as one of the conversion methods.

[0186] Tustin transformation: s = (2 / dt)·(z-1) / (z+1) (8)

[0187] Here, s is the Laplace operator, z is the shift operator, and dt is the discretized sampling time.

[0188] When expression (8) is used for organization instead of expression (4), expression (9) is established.

[0189] F(z,τ)=(n1+n0·z -1 ) / (1+d0·z -1 ) (9)

[0190] here,

[0191] n1=(2τ+dt) / (4τ+dt),

[0192] n0 = (-2τ + dt) / (4τ + dt), and

[0193] d0 = (-4τ + dt) / (4τ + dt).

[0194] at last, Figure 8 Medium target voltage V r With the correction value V r-fil The relationship between is represented by expression (10). Expression (10) can be calculated on the vehicle-mounted computer as shown in expression (11).

[0195] V r-fil =F(z,τ)V r =(n0+n0·z -1 )×V r / (1+d0·z -1 ) (10)

[0196] V r-fil [i] = -d0·V r-fil [i-1]+n0·V r [i]+n1·V r [i-1] (11)

[0197] Here, i represents discrete time, i = 1, 2, 3, ... The coefficients d0, n1, and n0 in expression (11) include the time constant τ, as shown in expression (9), and thus can be obtained according to the mapping diagram ( Fig. 9 , Fig.11 or Fig.12 ) to change the value of the coefficient.

[0198] [2.5. Handling multiple voltages and catalyst material changes]

[0199] The dynamic filter F(s, τ) of expression (4) corresponds only to the representative voltage V determined in step 2. Therefore, for all voltages V in step 2, the time constant τ in expression (4) is calculated and mapped in advance. Table 1 shows an example of a mapping diagram (mapping diagram A) of voltage V and time constant τ. By using mapping diagram A, the response speed (time constant τ) of the dynamic filter is adjusted according to the target voltage V. r Change.

[0200] [Table 1]

[0201] Voltage V[V] Time constant τ 0.60 <![CDATA[τ1]]> 0.65 <![CDATA[τ2]]> 0.70 <![CDATA[τ3]]> 0.75 <![CDATA[τ4]]> 0.80 <![CDATA[τ5]]> 0.85 <![CDATA[τ6]]> 0.90 <![CDATA[τ7]]>

[0202] Fig. 9 A schematic diagram showing a method for processing multiple voltages. For example, when V r When V is 0.70 V, τ3 is read out from map A. Then, a dynamic filter F(z, τ) is generated by using the read τ3. r After being input into the obtained dynamic filter F(z,τ), the output is the same as V r The corresponding correction target voltage V r-fil .

[0203] In addition, when V r When between 0.80 and 0.85 V, τ5 and τ6 are read from map A and used to linearly interpolate τ. The linearly interpolated τ is then used to generate the dynamic filter F(z, τ). r Input the obtained F(z,τ) and output the same as V r The corresponding correction target voltage V r-fil Therefore, a wide range of voltage V can be handled.

[0204] In addition, if the material of the catalyst particles changes, the time constant τ also changes. Therefore, for each material of the catalyst particles, the map A shown in Table 1 is prepared in advance. When the material of the catalyst particles changes, the time constant τ is changed by using the map A corresponding to the material. Therefore, various materials of the catalyst particles can be processed.

[0205] [2.6. Treatment of catalyst degradation]

[0206] It is known that when catalyst particles are used in an environment causing voltage fluctuations, a degradation phenomenon occurs over time, in which the number of particles with a small particle diameter decreases and the number of particles with a large particle diameter increases. When such degradation occurs, even if the voltage V does not change, the response speed (time constant τ) of the cover changes.

[0207] Fig.10 Schematic diagram showing the change in transfer characteristics when the catalyst degrades. Fig.10 As shown, in the new catalyst ( Fig.10 A) and degraded catalyst ( Fig.10 B) The time constant τ is different between them. If the time constant τ is estimated only by the voltage V without considering the degradation of the catalyst, the estimation accuracy of τ will decrease.

[0208] Therefore, the relationship between the degree of degradation of the catalyst particles after degradation (i.e., the voltage reduction ΔV) and the response speed (time constant τ') is obtained in advance. If it is considered that the obtained τ' is the result of shifting the time constant τ before degradation by a constant multiple due to degradation, the relationship τ'=k·τ is obtained. The correction gain k at this time is calculated and a map B of k is prepared in advance. Table 2 shows an example of a map B of the correction gain k. The response speed (time constant τ) of the dynamic filter is corrected based on the map B.

[0209] [Table 2]

[0210] Voltage reduction △V[V] Correction Gain 0 1 -0.02 <![CDATA[k1]]> -0.04 <![CDATA[k2]]> -0.06 <![CDATA[k3]]> -0.08 <![CDATA[k4]]> -0.10 <![CDATA[k5]]>

[0211] Fig.11 A schematic diagram showing a method for treating catalyst degradation. For example, when ΔV is -0.02 V at a certain time t, k1 is read from map B. Then, τ is corrected using the read k1, and a dynamic filter F(z,τ) is generated by using the corrected τ. V r Input the obtained F(z,τ) and output the correction target voltage V corresponding to △V r-fil .

[0212] In addition, when ΔV is between -0.06 and -0.08 V, k3 and k4 are read out from the map B, and k is linearly interpolated by using them. Then, τ is corrected by using the linearly interpolated k, and the dynamic filter F(z, τ) is generated by using the corrected τ. r Input the obtained F(z,τ) and output the correction target voltage V corresponding to △V r-fil . Therefore, catalyst degradation can be dealt with.

[0213] The degree of catalyst degradation can be grasped as a change in IV characteristics, for example. If degradation progresses, the voltage V that can be output at the same current I decreases. The relationship between the reduction ΔV and the response speed (time constant τ) is grasped in advance by simulation or bench testing. When the map B of Table 2 is installed on the vehicle controller, the voltage reduction ΔV can be calculated based on the voltage V and current I obtained from the vehicle sensor (see Fig.11 ).

[0214] [2.7. Dealing with temperature and humidity changes]

[0215] When the temperature T of the fuel cell changes, the response speed (time constant τ) changes. Therefore, a representative relationship between the temperature T and the response speed (time constant τ') after the temperature change is obtained in advance. If it is considered that the obtained time constant τ' is the result of the time constant τ before the temperature change being offset by a constant multiple, the relationship τ'=p·τ is obtained. The correction gain p is calculated and a mapping map C of p is prepared in advance. Table 3 shows an example of a mapping map C of the correction gain p. In Table 3, -50°C, -25°C, 0°C, 30°C, 60°C and 120°C are set as representative temperatures.

[0216] [Table 3]

[0217] Temperature T[℃] Correction Gain -50 <![CDATA[p1 <!-- 12 -->]]> -25 <![CDATA[p2]]> 0 <![CDATA[p3]]> 30 <![CDATA[p4]]> 60 <![CDATA[p5]]> 90 <![CDATA[p6]]> 120 <![CDATA[p7]]>

[0218] The same applies to humidity. That is, if the humidity RH of the fuel cell changes, the response speed (time constant τ) changes. Therefore, a representative relationship between the humidity RH and the response speed (time constant τ') after the humidity change is obtained in advance. If it is considered that the obtained time constant τ' is the result of the time constant τ before the humidity change being offset by a constant multiple, the relationship τ'=r·τ is obtained. The correction gain r is calculated and a mapping map D of r is prepared in advance. Table 4 shows an example of a mapping map D of the correction gain r. In Table 4, 0%RH, 25%RH, 50%RH, 75%RH and 100%RH are set as representative humidities.

[0219] [Table 4]

[0220] Humidity RH[%] Correction Gain 0 <![CDATA[r1]]> 25 <![CDATA[r2]]> 50 <![CDATA[r3]]> 75 <![CDATA[r4]]> 100 <![CDATA[r5]]>

[0221] Fig.12 Schematic diagram showing a method for processing temperature changes and / or humidity changes. The temperature T is measured and a correction gain p corresponding to the temperature T is read from a map C. Alternatively or in addition, the humidity RH is measured and a correction gain r corresponding to the humidity RH is read from a map D. Subsequently, τ is corrected by using the read p and / or r, and a dynamic filter F(z,τ) is generated by using the corrected τ. V rThe obtained dynamic filter F(z,τ) is input, and the corrected target voltage V corresponding to △V is output. r-fil . Therefore, a wide range of temperature and humidity conditions can be handled.

[0222] [3. Effect]

[0223] Fig.13 A schematic diagram showing the voltage change during the transient time. Fig.13 As shown, by the simple configuration of the dynamic filter in the present invention, the rate of change of the rising portion of the catalyst voltage is further slowed down compared with the prior art. As a result, it is possible to prevent Figure 1 The degradation index shown temporarily increases (hatched area), and thus the degradation of the catalyst can be suppressed.

[0224] Meanwhile, the delay of the voltage value causes the fuel efficiency to deteriorate, but in the present invention, the filter is configured to minimize the voltage delay (in other words, only suppress Figure 1 ), so the degradation of fuel efficiency can be minimized. Therefore, in the present invention, it is possible to generate a voltage value in which the degradation of fuel efficiency is suppressed to a minimum while the degradation of the catalyst is suppressed.

[0225] Furthermore, although an example in which the target voltage is changed stepwise has been shown so far, the present invention can also be applied to a target voltage that is changed arbitrarily.

[0226] The embodiments according to the present invention have been described in detail so far, but the present invention is not limited to the above-described embodiments at all, and various changes can be made within the scope not departing from the gist of the present invention.

[0227] Industrial Applicability

[0228] The catalyst degradation suppressing device according to the present invention can be used for output control of a fuel cell system having various fuel cells.

Claims

1. A catalyst degradation suppression device, comprising: first means for obtaining a voltage V of a fuel cell as a variable for estimating a response speed at which a coverage ratio of an oxide film of catalyst particles contained in a cathode of the fuel cell changes, wherein the voltage V of the fuel cell is equal to the catalyst voltage V cat , and wherein the response speed is represented by a time constant τ; The second device is used to read the time constant τ corresponding to the voltage V at the current time t from the pre-made map A. t , the prefabricated map A represents the relationship between the voltage V and the time constant τ and corresponds to the catalyst particles; The third means is for using the time constant τ t Generate a continuous-time dynamic filter F(s,τ), and convert the continuous-time dynamic filter F(s,τ) into a discrete-time dynamic filter F(z,τ); and The fourth device is used to input the target voltage V to the discrete time dynamic filter F(z,τ) r And output the correction target voltage V r-fil .

2. The catalyst degradation suppression device according to claim 1, wherein the dynamic filter F(s, τ) is represented by the following expression (4) or (6), F(s, τ) = (τ·s + 1) / (2 τ·s + 1) (4) Here, s is the Laplace operator, F(s, τ) = G m (s) / G(s, τ) (6) Here, G m (s) is the transfer function of the ideal response model, and G(s,τ) is the transfer function of the model.

3. The catalyst degradation suppression device according to claim 1, wherein: The first device further comprises: (a) Device A, for obtaining the terminal voltage V of the fuel cell cell , current I and resistance R, and according to the relational expression V = V cell +IR to calculate the voltage V; (b) Device B, for obtaining V cell , V cell Substitute the previously obtained cell The relationship between the voltage V and V = f(V cell ), and estimating the voltage V; or (c) A device C for installing a reference electrode in advance on the cathode-side catalyst layer in the fuel cell and actually measuring the voltage V by using the potential difference between the reference electrode and the cathode-side current collector.

4. The catalyst degradation suppression device according to claim 1, in, The first means comprises means for obtaining a voltage decrement ΔV as the variable; The second device comprises a device for reading out the correction gain k corresponding to the voltage reduction ΔV at the current time t from the pre-made map B t , and multiplying the time constant τ by the correction gain k t The prefabricated map B represents the relationship between the voltage reduction ΔV and the correction gain k of the time constant τ and corresponds to the catalyst particles; and The third device includes a device for using the correction gain k t The device generates the continuous-time dynamic filter F(s, τ) by using the corrected time constant τ, and converts the continuous-time dynamic filter F(s, τ) into the discrete-time dynamic filter F(z, τ).

5. The catalyst degradation suppression device according to claim 1, in, The first means comprises means for obtaining the temperature T of the fuel cell as the variable; The second device comprises a device for reading out the correction gain p corresponding to the temperature T at the current time t from the pre-made map C. t , and multiplying the time constant τ by the correction gain p t The prefabricated map C represents the relationship between the temperature T and the correction gain p of the time constant τ and corresponds to the catalyst particles; and The third device includes a device for using the correction gain p t The device generates the continuous-time dynamic filter F(s, τ) by using the corrected time constant τ, and converts the continuous-time dynamic filter F(s, τ) into the discrete-time dynamic filter F(z, τ).

6. The catalyst degradation suppression device according to claim 1, in, The first means comprises means for obtaining, as the variable, the humidity RH of the gases respectively supplied to the cathode and anode of the fuel cell; The second device comprises a device for reading out the correction gain r corresponding to the humidity RH at the current time t from the pre-made map D. t , and multiplying the time constant τ by the correction gain r t The prefabricated map D represents the relationship between the humidity RH and the correction gain r of the time constant τ and corresponds to the catalyst particles; and The third device includes a device for using the correction gain r t The device generates the continuous-time dynamic filter F(s, τ) by using the corrected time constant τ, and converts the continuous-time dynamic filter F(s, τ) into the discrete-time dynamic filter F(z, τ).

Images