• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to a method for determining critical operating states on a fuel cell stack consisting of series-connected individual cells, wherein the fuel cell stack is impressed with a low-frequency current or voltage signal, the resulting voltage or current signal is measured and the harmonic distortion thd is determined. According to the invention, the weighted sum of a temperature dependent on the membrane resistance Rm and a temperature dependent on the total harmonic distortion THD is used to determine an indicator THDAdryout correlating with the drying out of the fuel cell membranes of the fuel cell stack, the membrane resistance Rm being detected by impedance measurement.
    公开号:AT512888A1
    申请号:T50154/2012
    申请日:2012-05-03
    公开日:2013-11-15
    发明作者:
    申请人:Avl List Gmbh；
    IPC主号:
    专利说明:

    The invention relates to a method for determining critical operating states on a fuel cell stack consisting of series-connected individual cells, wherein the fuel cell stack a low-frequency current or voltage signal impressed, the resulting voltage or current signal measured and the harmonic distortion of the measured signal is determined.
    In the production of fuel cells, quality assurance requires a review of the functionality or performance of all cells. This is done according to the prior art, for example by means of measurement of the individual cell voltages. However, due to the high cost, the measurement of single cell voltages is not desired either during fabrication or during operation of fuel cells.
    In the laboratory so-called impedance spectroscopy is also used to detect the operating status or "health status". used in the fuel cell stack. In doing so, the complex impedance (i.e., impedance locus) of the fuel cell stack is measured over a particular frequency range and compared, for the most part, with reference curves.
    Depending on which frequency sets typical changes in the impedance curve, it can now be distinguished whether these changes from the anode, the cathode or the membrane of the individual cells emanate. The method is based on the fact that the electrical equivalent circuit diagram for the fuel cell stack comprises a series-parallel connection of low-pass elements 1.
    Is order whose cut-off frequencies are significantly far apart and therefore the desired selectivity can be achieved.
    In essence, there are the following effects in the fuel cell stack (for example, in a PEM fuel cell operated with air and H2) that require monitoring.
    Oxidizer or denier undersupply at cathode or anode (substoichiometry). Effect: U / I characteristic drops even at lower currents.
    2
    Membrane: formation of electrical short circuits or gas short circuits. Effect: U0 (voltage at current = 0) changes.
    Electrode aging. Effect: U / I characteristic decreases steeper, greater ohmic resistance due to corrosion effects.
    A disadvantage of pure impedance spectroscopy is the relatively large measurement effort, in addition, the impedance spectroscopy is time-consuming, since at each of the gradually increasing frequencies, the impedance in the steady state must be measured.
    Benefits include the Total Harmony Distortion Analysis (THDA), an online diagnostic tool for determining the status of fuel cell stacks. With relatively little measurement effort, parameters can be extracted which can be used for the further calculations of the state variables of the fuel cell stack.
    Such a method based on the analysis of the distortion factor is described in detail in EP 1 646 101 B1, in which the fuel cell stack is impressed with a low-frequency current or voltage signal, the resulting voltage or current signal is measured and at least one change of the harmonic content (or the distortion factor) of the signal on the operating state of individual lines of the fuel cell stack is closed. This allows an online evaluation of the above-mentioned measurements relating to system level detection and classification of critical operating conditions.
    Harmonic analysis can be done either in the time domain by using filters (digital or analog filters) or by transforming them into the frequency domain (using all kinds of wavelet transforms, short-time Fourier transforms or fast Fourier transforms). The advantage of the frequency transformation is that the signal / noise ratio is significantly improved by this transformation, which in turn increases the sensitivity of the measurement method.
    The object of the invention is based on the method known from EP 1 646 101 Bl, based on the analysis of the harmonic distortion 3
    To propose method variants with which different critical operating states of fuel cell stacks can be clearly detected, such as: • Sub stoichiometry at anode / cathode of the stack • Drying of the membranes of the stack • Water accumulation, droplet formation at the membranes • Decline of the cell voltage of single cells of the stack
    A first solution according to the invention provides that the weighted sum of a term dependent on the membrane resistance Rm and a term dependent on the harmonic distortion thd is used to determine an indicator correlating with the drying out of the fuel cell membranes of the fuel stack. The membrane resistance Rm is detected by impedance measurement.
    A second solution according to the invention provides that the weighted sum of a term dependent on the internal resistance Ri, a term dependent on the harmonic distortion thd, and a value dependent on the stoichiometric undersupply of the anode and / or cathode side of the fuel cell stack is THDA | OW media the impedance Rim of the low-frequency signal dependent term is used.
    A third solution according to the invention provides that the parameters thddiio and thddifi and the fluctuations fd (V) of the measured voltage curve are used to determine an indicator THDA | iC | Uid which correlates with impermissible accumulations of water and droplet formation on the membranes of the fuel stacks. Thddifo and thddifi are each a linear combination of the distortion factors of current and voltage, where thddm comprises the portion of the first harmonics and thddin the portion of the second harmonics.
    The indicator THDAdryom gives a percentage indication of the dryness status of the membranes in the stack. THDAiM ^^ returns the degree of 4
    Media undersupply at the cathode or the anode (media here include air, hydrogen, or methanol). THDAnqU | d indicates the occurrence of unwanted water retention.
    According to a first advantageous development of the invention, a simplified electrical equivalent circuit diagram of the fuel cell stack can be used to determine a correlating with the aging of the fuel cell stack indicator SoH, at least the ohmic resistances of the cathode side and the anode side Ri, R2, and the cathode side and anode side double-layer capacitances Ci, C2 and the inductance Lm taken into account, wherein from impedance measurements at least three measurement frequencies, the parameters for a Gieichungssystem for the variables to be determined Ri, R2 / Ci, C2 / Lm is set up, and its solution is at least partially used to calculate the indicator SoH.
    In this case, three measurement frequencies are preferably selected for the calculation in which the impedance curve of the simplified equivalent circuit diagram substantially coincides with the real impedance curve of the fuel cell stack.
    According to a second advantageous development of the invention, an artificial neural network (ANN) can be used to determine an indicator avg-min which correlates with the minimum of the cell voltage of an individual cell of the fuel cell stack, with THDA derived from the harmonic distortion analysis as input variables of the neural network Measured variables and derived from the real and imaginary part of the impressed current and voltage signal derived impedance values are used, and wherein the neural network is trained to determine the internal network parameters by means of signals from single cell voltage measurements.
    Preferably, a two-layer feed-forward artificial neural network FF ANN can be used to simulate the indicator avg-min correlating with the minimum of the cell voltage of a single cell of the fuel cell stack, the neural network having a training function, preferably the Levenberg-Marquardt training function, to which Single-cell voltage measurement recorded measured values is adjusted. 5
    The invention is explained in more detail below with reference to schematic representations. Show it:
    1 is an equivalent circuit diagram of a fuel cell stack,
    2 is a schematic sectional view of a fuel cell,
    3 is a schematic of a two-layer Feed Forward Artificial Neural Network, as well
    4 is a schematic of a functional set of a THDA diagnostic tool. THDAdryout
    The decisive parameter for THDAdryout is the membrane resistance Rn, (see Fig.l). Depending on the adaptation of certain system-dependent parameters (a0, al}
    Reference value ref as a function of the age of the stack) results in the calculation of THDAdryout a polynomial depending on the membrane resistance Rn ,. It can be seen that increasing membrane resistance is an indication of increased membrane dehydration. Furthermore, the distortion factor (thd, ratio of the fundamental to its harmonic content) serves as an additional indicator of the non-linear behavior of the system response. K ~ ref ref + ax fifhd)
    formula 1
    With system-dependent weights α0, α, < = R, where or0 + or, = 1, and polynomial or logarithm function f.
    The membrane resistance R 1 can be extracted from the highest frequency of the impressed signal since the equivalent circuit of the fuel cell stack is a series-parallel circuit of first order low-pass elements whose cut-off frequencies are significantly far and much lower. 6 THDA | OW medta
    Insufficient media supply is detected by means of three parameters. Both the internal resistance Ri and the harmonic distortion factor thd of the voltage signal of the system response to the impressed signal and an impedance R | m of the low-frequency signal play an important role here. THDAlowmedia -αγ · fx (thd) + a2-f2 (R,) + a3 / 3 (Rlm) Formula 2
    With system dependent weights ax, a2, a3 e R where ax + a2 + a3 = 1 and
    Evaluation functions fi, f2, fß. The internal resistance Ri is the sum of the ohmic resistances Rm, Ri and R2 and is calculated online using the following formula: V -V R,
    ideally,
    Where V is the stack voltage, I is the stack current, V0 is the open circuit voltage, and n "is the number of single cells in the stack. THDA | Quid
    The appearance of accumulation of water on the membranes can be observed by examining the harmonic distortion. Rapid small fluctuations in the voltage curve also serve as an indicator. The formula is THDAliquld = a0 · / («! · Abs (thdm) + cc2 · α & (ί / κ / ώ / 1)) + a3 · fd (v) Formula 3
    Where e R are system-dependent parameters, fine polynomial or one
    Logarithm function, and thddiro or thddin is a linear combination of the distortion factors of current and voltage of the two measured channels (current, voltage).
    In order to suppress distortion in harmonic distortion, as soon as there are disturbances in the current and tension distortion factor, the difference (thd <nio, thddifi) of these 7 are formed. Furthermore, the two lowest frequencies of the impressed signal are preferably selected very close to each other, since in the case of a distortion of the distortion factor, the other (interference-free) frequency can be used.
    Optionally, the term fd (V), the finite difference of the voltage values, since these values can also clearly indicate the occurrence of water droplets, depending on the fuel cell type.
    State of Health (SoH) - Aging
    The result of the state of health measurement reflects the degree of aging of the stack. Here, a new stack is assigned 100% SoH and an end-of-life (e.g., 90% performance loss) 0% SoH. By means of impedance measurement and a simplified electrical equivalent circuit diagram (see FIG. 1), the following equations can be established. z «: - -Vl1 - Ä] + - &quot; v! 1 - Ä l + 'U1 + -A1 1 + Ω, 1 + Ω * with Ω, = - g> 2R2C2, Ω, = <y3Z, m
    Where α> ι, ω2, and ω3 are the corresponding angular frequencies.
    Considering the real and imaginary parts of the above complex impedance separately, this yields: MZn:} = - ^ T + - ^ + ^, and lm {zrcj = --½¾ + Λ "Ω, 1 fC &gt; i + a * ι + ω22 * 1 i + ω ι + Ω *
    A simplification of these formulas leads to the following equations:
    Re {} = ZFC - or. lm {ZfC} * - 1 FCi i + qJ i + al * 1 FCi i + qJ 1 + Ω2 for frequencies between 5 and 10 Hz.
    Re {z ^}! R, 1 + Ω: + Rm or Im {z ^} * R,, Ω, 1 '"bi" 3 δ for frequencies between 10 and 100Hz and
    Re (Zfc) Ä Rjh "TT - pr- + J ^ m ^ 3
    Lij ll2 for frequencies over 400Hz.
    These equations are set up for three frequencies and the corresponding real and imaginary parts of the impedances and resolved according to Ci, C2, U ,, Ri and R2. Surprisingly, it has been found that the aging correlates clearly and approximately linearly with the two double-layer capacitances Ci (cathode side) and C2 (anode side) and with the ohmic resistances of cathode (Ri) and anode (R2).
    By numerically solving the system of equations, one obtains current values for the individual components. For the following practical reasons, the ohmic resistances are not used for the aging measurement, but only the double-layer capacitances: too many side effects, i. E. Ohmic resistances also correlate strongly with operating parameters such as pressure, temperature and media stoichiometry. - To reliably determine the ohmic resistances, very small frequencies (for example below 1 Hz) are necessary. Disadvantage: too long measuring times and significant disturbing influences which derive from the load dynamics.
    Consequently, only Ci and C2 are used to calculate the SoH value. The following formula is used for this:
    SoH - oii 0.75Ct 3 + flf, 100--4 = - 100 0.75C2 3;
    Formula 4
    With system dependent weights aua2 € R, where ¢ ^ + 0 ^ = 1 and the starting values Cj and C2 of the new fuel cell stack.
    To minimize the error of the aging calculation by means of a simplified equivalent circuit diagram instead of known complex and exactly matched equivalent circuit diagrams, an advantageous method is used. The three measurement frequencies are selected so that the impedance curve of the simplified 9
    Equivalent circuit diagram with the real impedance curve of the fuel cell exactly at these three frequencies is congruent.
    To ensure the reliability of the above formulas, the quantities used are smoothed as needed by using weighted weighted averaging. This corresponds to a filter that damps high frequencies.
    2 shows a schematic representation of the section through a membrane-electrode unit MEA of a PEM fuel cell with a membrane 11 lying between an anode 10 and a cathode 11. The anode and cathode 10, 12 each consist of a diffusion layer 13 and a catalytic layer 14, 15. The fuel supply, for example H2, at the anode is indicated at 16, the outlet at 16 '. The supply of the oxidizing agent (for example air) takes place at 17, the removal at 17 '.
    Minimum cell voltage avg-min
    The invention further relates to a method for the numerical determination of the deviation of a cell voltage minimum from the average cell voltage in a fuel cell stack. This method serves as a replacement for a single cell voltage meter.
    To monitor a fuel cell stack, it is common practice to perform single cell voltage measurements. A Single Cell Voltage Meter (CVM) makes it possible to monitor the voltages of each fuel cell in a stack. It is particularly important to recognize whether the voltage of individual cells drops. This can no longer be read with a large number of cells from the total voltage of the stack (sum of all single cell voltages). Often certain incidents (critical states) initially affect a few cells before there is a significant drop in the total voltage. To detect this, the deviation of the cell voltage minima avg-min from the average cell voltage must be observed or determined. This size thus provides the deviation of the smallest single cell voltage from the average cell voltage. The 10
    However, single-cell voltage monitoring is time-consuming and a costly and space-intensive method and thus only conditionally suitable for series production.
    The method according to the invention allows the monitoring of the avg-min signal without resorting to the single-cell voltage measurement data. It is a process based on the THDA process. By impressing a modulated current or voltage signal can be concluded by means of measurement and the corresponding analysis of the system response (voltage or current) on the operating condition of the fuel cell stack. As described above, the analysis provides three characteristics THDAdryom, THDA | OWme <iia and THDAitquid for state detection, as well as phase and amplitude of the impedances at different frequencies. Furthermore, the harmonic distortion factor thd of the system response is also calculated. The procedure described below uses all measures from the THDA (real and imaginary parts of impedances, distortion factors, SNR (Signal to Noise Ratio)) to calculate the avg-min signal.
    The data analysis and algorithm development is based on an artificial neural network (ANN). It is a network of artificial neurons, an abstraction of natural brain neurons used in information processing and artificial intelligence. The areas of application for ANN are diverse and range from function fitting and classification problems to pattern recognition or time series analysis. If the architecture of the network is fixed, it is trained by means of appropriate training algorithms, that is, weights and parameters inside the network are adapted.
    To simulate the avg-min signal, a two-layer feed-forward ANN (FFANN) with, for example, 10 neurons in the hidden layer was used (see FIG. 3). All available measures of the THDA method provide the inputs for the neural network, with the Levenberg-Marquardt training function, the network is trained on the fit of the original (measured by CVM meter) avg-min signal.
    Since the result of a network trained with the entire data set is not always reliable, the training data set S is divided with respect to a threshold value T. The division is based on the avg 11 min signal measured in the stack for test purposes in two data sets. The first set, Si contains all the input quantities that an avg-min-signa! &Lt; T and for the second set S2 = S
    Sl. Since many systems switch off when they fall below a minimum value of the single-cell voltage, or are in a critical state, this division of the data into two groups can also be classified as "uncritical". and &quot; critical &quot; Interpret situations. The choice of the threshold is system dependent and the method for online separation of the data is described in paragraph 8 described in more detail. The reason for the split is that the corresponding avg-min values can be simulated better if one FFANN per data set is trained. The result (out) of a FFC constructed for this process can be described by the formula: outiin) = f2 (W2 /, (WL -in + b ^) + b2) Formula 5
    Here, in is the data input vector and in fi, f2 is transfer functions. The weight matrices Wi, W2 and the blow vectors bi, b2 are optimized accordingly during the training.
    The data separation results in two FFANNs Nt, and N2, which are trained with the corresponding amounts of data and can be described by two mathematical functions outi and out2. These two new functions expand the function set of the THDA diagnostic tool (see Fig. 4).
    Using standard classification algorithms (Support Vector Machine, PCA, Nearest Neighbor, Cluster Analysis), the data is separated online with regard to a system-dependent threshold value. An observation, consisting of one instance of each measured value, is classified accordingly and then provides the input quantity for the respective neural network. Depending on the results that these networks provide, the degree of criticality of the fuel cell stack can be estimated and additional steps such as control measures taken.
    权利要求:
    Claims (13)
    [1]
    1. A method for determining critical operating states on a fuel cell stack, consisting of series-connected individual cells, wherein the fuel cell stack a low-frequency current or voltage signal impressed, the resulting voltage or current signal measured and the harmonic distortion thd is determined, characterized in that for determining an indicator THDAoryout which correlates with the drying out of the fuel cell membranes of the fuel cell stack, the weighted sum of a term dependent on the membrane resistance Rm and a term dependent on the harmonic distortion thd is used, the membrane resistance Rm being detected by impedance measurement.
    [2]
    2. The method according to claim 1, characterized in that the correlating with the drying of the fuel cell membranes indicator THDAdryout according to

    where a0 + at = 1 and ref is a reference value for the membrane resistance and f is a polynomial or a logarithmic function.
    [3]
    3. A method for determining critical operating conditions on a fuel cell stack consisting of series-connected single cells, wherein the fuel cell stack impressed a low-frequency current or voltage signal, the resulting voltage or current signal measured and the harmonic distortion thd of the signal is determined, characterized in that for determining an indicator THDA ^ media correlating with the stoichiometric undersupply of the anode and / or cathode side of the fuel cell stack, the weighted sum of a term dependent on the internal resistance Ri, a term dependent on the harmonic distortion thd, and a term dependent on the impedance Rim of the low-frequency signal is used becomes. 13
    [4]
    4. The method according to claim 3, characterized in that the correlating with the stoichiometric undersupply of the anode and / or cathode side of the fuel cell stack indicator THDAiowmedta according to THDAlovmidia = ay-fx (thd) + or2 / 2) + αν / 3 (¾.) where ax + a2 + a3 = 1 and fl and f2 are polynomials or logarithmic functions.
    [5]
    5. A method for determining critical operating conditions on a fuel cell stack, consisting of series-connected single cells, wherein the Brennstoffzeilenstack a low-frequency current or voltage signal impressed, the resulting voltage or current signal measured and the harmonic distortion thd of the signal is determined, characterized in that for determining an indicator THDAnqU | d correlating with impermissible accumulations of water and droplet formation on the membranes of the fuel cell stack, the parameters thddin &gt; and thddm, as well as the fluctuations fd (V) of the measured voltage curve, where thddm and thddifi are each a linear combination of the distortion factors of current and voltage.
    [6]
    6. Method according to claim 5, characterized in that the indicator THDAuqUid, which correlates with impermissible accumulations of water and droplet formations on the membranes of the fuel cell stack, is determined according to mDAti <, * d = "ο 'fiai" bs (thddifμ) + a2' βΨ), where a0, aXia2> a3 are system dependent parameters.
    [7]
    7. The method according to any one of claims 1 to 6, characterized in that for additional determination of a correlating with the aging of the fuel cell stack indicator SoH a simplified electrical equivalent circuit diagram of the fuel cell stack is used, the at least the ohmic resistances of the cathode side and the anode side Ri, R2 / and the cathode-side and anode-side 14 double-layer capacitances Ci, C2 taken into account, wherein a system of equations for the parameters Ri, R2 / Cif C2 is set up, which are determined by impedance measurements at at least three measurement frequencies and used to calculate the indicator SoH.
    [8]
    8. The method according to claim 7, characterized in that the correlation with the aging of the fuel cell stack indicator SoH from the parameters Ci, C2 of the double-layer capacitances according to SoH = ax 100- 0.75Q 100 + αΊ 100-- 0.75C, 100 3 Λ is determined , wherein the parameters the ohmic resistances of the cathode side and the anode side Ri, R2 are disregarded, + a2 = 1, and C, and C2 are start values of a new fuel cell stack.
    [9]
    9. The method according to claim 7 or 8, characterized in that three measurement frequencies are selected for the calculation in which the impedance curve of the simplified equivalent circuit diagram with the real impedance curve of the fuel cell stack is substantially the same.
    [10]
    10. The method according to claim 2 and 4, characterized in that a stoichiometric undersupply of the anode side of the fuel cell stack by the combined occurrence of an increasing membrane resistance R », according to indicator THDA4ry0Ut and a deviation of the internal resistance R | from the reference value according to indicator THDAiowmedia.
    [11]
    11. The method according to claim 4 and 6, characterized in that a stoichiometric undersupply of the cathode side of the fuel cell stack by the combined occurrence of an increasing harmonic distortion thd according to indicator THDA | iqU) (i and a deviation of the internal resistance Ri is determined by the reference value according to indicator THDAiowmecna. 15
    [12]
    12. The method according to any one of claims 1 to 6, characterized in that for the additional determination of a correlating with the minimum of the cell voltage of a single cell of the fuel cell stack indicator avg-min an artificial neural network (Artificial Neural Network) ANN is used, that as input variables of Neural network derived from the harmonic distortion THDA derived measurements and derived from the real and imaginary part of the impressed current and voltage signal derived impedance values are used, the neural network is trained to determine the internal network parameters using signals from single cell voltage measurements.
    [13]
    13. The method according to claim 12, characterized in that for simulating the correlating with the minimum of the cell voltage of a single cell of the fuel cell stack indicator avg-min a two-layer Feed-Forward Artificial Neural Network FFANN is used and the neural network with a training function, preferably the Levenberg -Marquardt training function to which readings taken by single cell voltage measurement is adjusted. 2012. 05 03 Lu / St
    类似技术:
    公开号 | 公开日 | 专利标题
    AT512888B1|2014-11-15|Method for determining critical operating states on a fuel cell stack
    EP1646101B1|2008-08-13|Method of monitoring the operational state of a fuel cell stack
    DE10303440A1|2003-09-04|Moisture sensor and a fuel cell system that uses it
    DE112006002187T5|2008-06-12|Fuel cell system and operating method for the fuel cell system
    DE102008047389A1|2009-04-30|Online system determination and control of fuel cell humidification via HFR measurements
    DE102010005644A1|2010-11-18|A system and method for monitoring an anode fluid composition during a fuel cell system startup
    DE102016111437A1|2017-06-14|A method and apparatus for diagnosing a condition of a fuel cell stack
    DE10036572A1|2002-02-14|fuel cell plant
    EP3443610A1|2019-02-20|Fuel cell forecast model based on an equivalent circuit diagram
    EP2546646B1|2016-11-16|Analysis of gases dissolved in transformer oil
    EP3209412A1|2017-08-30|Checking the tightness of at least one membrane of an electrolyzer
    DE102015225168A1|2016-06-16|Apparatus and method for measuring impedance for diagnosis of a fuel cell
    DE102007040837A1|2009-03-05|Method for non-destructive recognition of leakage gas in fuel cell operated with oxygen and hydrogen, involves arranging ion-permeable electrolyte between anode gas area and cathode gas area
    DE102012200026A1|2013-07-04|Method for e.g. determining concentration of gas constituent in exhaust gas of internal combustion engine, involves determining temperature based on computed quotients and predetermined correlation between quotients and temperature
    DE102011084734A1|2013-04-18|Method for adjusting a gas sensor
    DE102015201056A1|2016-07-28|Sensor arrangement for determining a hydrogen content of a sample gas and corresponding method
    DE102008023183A1|2009-11-12|A method of detecting and locating defects in a fuel cell within a fuel cell stack
    DE10148855A1|2003-04-17|Sensor comprises a proton-conducting electrolyte arranged between catalytically active electrodes, and a measuring unit for evaluating a temporary change of a first electrical operating parameter of the sensor
    WO2018083147A1|2018-05-11|Method for diagnostics of a technical system
    同族专利:
    公开号 | 公开日
    CN106684405A|2017-05-17|
    CN107064807B|2019-06-21|
    EP3285322A1|2018-02-21|
    WO2013164415A1|2013-11-07|
    US10145900B2|2018-12-04|
    US20190049522A1|2019-02-14|
    US10416240B2|2019-09-17|
    KR101969741B1|2019-04-17|
    CN106684405B|2019-06-18|
    JP6195905B2|2017-09-13|
    US10345389B2|2019-07-09|
    CN104396073A|2015-03-04|
    US20150153418A1|2015-06-04|
    EP2845255A1|2015-03-11|
    EP3285321B1|2019-03-27|
    US20190049521A1|2019-02-14|
    EP3285321A1|2018-02-21|
    CN104396073B|2017-03-15|
    CN107064807A|2017-08-18|
    EP2845255B1|2017-10-04|
    AT512888B1|2014-11-15|
    KR20150013224A|2015-02-04|
    EP3285322B1|2019-03-27|
    JP2015517714A|2015-06-22|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    DE10036572A1|2000-07-27|2002-02-14|Bosch Gmbh Robert|fuel cell plant|
    EP1646101B1|2004-10-07|2008-08-13|AVL List GmbH|Method of monitoring the operational state of a fuel cell stack|
    WO2006056076A1|2004-11-29|2006-06-01|Hydrogenics Corporation|Systems and methods for detecting and indicating fault conditions in electrochemical cells|
    EP1923945A1|2005-08-09|2008-05-21|Toyota Jidosha Kabushiki Kaisha|Device and method for judging deterioration in performance|
    US20120038452A1|2009-02-24|2012-02-16|Helion|Method for determining a state of health for an electrochemical device|
    DE102011112998A1|2010-09-15|2012-04-05|Gm Global Technology Operations Llc, |A low cost method and signal processing algorithm for rapidly detecting abnormal operation of a single fuel cell in multiple series connected fuel cells|
    US6456988B1|1997-03-12|2002-09-24|U.S. Nanocorp Inc.|Method for determining state-of-health using an intelligent system|
    JP2001076752A|1999-07-06|2001-03-23|Matsushita Electric Ind Co Ltd|Inspection device for and inspection method of battery element|
    JP2004510312A|2000-09-28|2004-04-02|プロトン　エネルギー　システムズ，インク．|Regenerative electrochemical cell system and operating method thereof|
    US6696190B2|2001-06-29|2004-02-24|Plug Power Inc.|Fuel cell system and method|
    US20030184307A1|2002-02-19|2003-10-02|Kozlowski James D.|Model-based predictive diagnostic tool for primary and secondary batteries|
    JP2006024437A|2004-07-07|2006-01-26|Matsushita Electric Ind Co Ltd|Service life estimation method for fuel cell system, and operating method for fuel cell system|
    JP4930818B2|2004-07-09|2012-05-16|トヨタ自動車株式会社|Fuel cell system|
    JP5017783B2|2005-03-09|2012-09-05|トヨタ自動車株式会社|Fuel cell system|
    JP5146898B2|2005-08-10|2013-02-20|トヨタ自動車株式会社|FUEL CELL POWER CONTROL DEVICE, FUEL CELL SYSTEM, AND FUEL CELL POWER CONTROL METHOD|
    JP4868240B2|2007-05-10|2012-02-01|トヨタ自動車株式会社|Fuel cell system|
    US8111037B2|2008-06-27|2012-02-07|GM Global Technology Operations LLC|Method for battery state-of-health monitoring using battery voltage during vehicle starting|
    FI20085976A|2008-10-17|2010-04-18|Waertsilae Finland Oy|Fuel cell arrangements comprising fuel cell stacks|
    JP5387172B2|2009-06-30|2014-01-15|日産自動車株式会社|Fuel cell internal state detection device|
    US8206860B2|2009-10-09|2012-06-26|GM Global Technology Operations LLC|Method to perform adaptive voltage suppression of a fuel cell stack based on stack parameters|
    CN201828651U|2010-10-26|2011-05-11|上海天晶电力科技发展有限公司|Digital full function state monitoring and safety evaluation device of transformer substation direct current system|
    KR101090705B1|2010-11-25|2011-12-08|강남대학교 산학협력단|Method for monitoring of fuel cell stack status|US10279702B2|2013-12-17|2019-05-07|Hyundai Motor Company|Technique of diagnosing fuel cell stack|
    KR101592641B1|2013-12-17|2016-02-12|현대자동차주식회사|Method for diagnosing fuel cell stack|
    KR101704224B1|2015-06-26|2017-02-07|현대자동차주식회사|Device and method for controlling fuel cell system|
    EP3133411B1|2015-08-18|2018-12-19|AVL List GmbH|Method for monitoring the operational condition of fuel cells|
    KR20170085365A|2016-01-14|2017-07-24|삼성전자주식회사|Method and apparatus for estimating state of battery|
    DE102016106735A1|2016-04-12|2017-10-12|Thyssenkrupp Marine Systems Gmbh|Replacement Switch Based Fuel Cell Prediction Model|
    JP6686920B2|2017-02-01|2020-04-22|株式会社Ｓｏｋｅｎ|Fuel cell system|
    US10249893B2|2017-04-26|2019-04-02|GM Global Technology Operations LLC|Fuel cell architectures, monitoring systems, and control logic for characterizing fluid flow in fuel cell stacks|
    JP6543671B2|2017-09-29|2019-07-10|本田技研工業株式会社|Fuel cell output inspection method|
    AT522011A3|2018-12-20|2021-03-15|Avl List Gmbh|Operating device, fuel cell system, motor vehicle and method for operating a fuel cell system|
    WO2021040146A1|2019-08-27|2021-03-04|Samsung Electronics Co., Ltd.|Method and system for determining health parameter of a battery|
    CN111007404A|2019-12-24|2020-04-14|萱柯氢能科技有限公司|Fuel cell impedance measurement and analysis system and method based on key frequency points|
    DE102020106563A1|2020-03-11|2021-09-16|Audi Aktiengesellschaft|Method for operating a motor vehicle with a fuel cell device and a motor vehicle|
    DE102020112916A1|2020-05-13|2021-11-18|Audi Aktiengesellschaft|Method for operating a fuel cell device, fuel cell device and motor vehicle with such|
    DE102020124075A1|2020-09-16|2022-03-17|Audi Aktiengesellschaft|Determination of the membrane resistance of a fuel cell|
    法律状态:
    优先权:
    申请号 | 申请日 | 专利标题
    ATA50154/2012A|AT512888B1|2012-05-03|2012-05-03|Method for determining critical operating states on a fuel cell stack|ATA50154/2012A| AT512888B1|2012-05-03|2012-05-03|Method for determining critical operating states on a fuel cell stack|
    PCT/EP2013/059172| WO2013164415A1|2012-05-03|2013-05-02|Method for determining critical operating states in a fuel cell stack|
    CN201710063564.3A| CN107064807B|2012-05-03|2013-05-02|Method for determining the critical operating state in fuel cell unit|
    EP13722342.6A| EP2845255B1|2012-05-03|2013-05-02|Method for determining critical operating states in a fuel cell stack|
    EP17193193.4A| EP3285322B1|2012-05-03|2013-05-02|Method for determining critical operating states on a fuel cell stack|
    US14/398,038| US10145900B2|2012-05-03|2013-05-02|Method for determining critical operating states in a fuel cell stack|
    CN201710063565.8A| CN106684405B|2012-05-03|2013-05-02|Method for determining the critical operating state in fuel cell unit|
    EP17193144.7A| EP3285321B1|2012-05-03|2013-05-02|Method for determining critical operating states on a fuel cell stack|
    KR1020147033872A| KR101969741B1|2012-05-03|2013-05-02|Method for determining critical operating states in a fuel cell stack|
    JP2015509443A| JP6195905B2|2012-05-03|2013-05-02|Method for measuring the critical operating state of a fuel cell stack|
    CN201380032641.9A| CN104396073B|2012-05-03|2013-05-02|Method for determining the critical operating state in fuel cell unit|
    US16/164,375| US10345389B2|2012-05-03|2018-10-18|Method for determining critical operating states in a fuel cell stack|
    US16/163,852| US10416240B2|2012-05-03|2018-10-18|Method for determining critical operating states in a fuel cell stack|
    [返回顶部]