CN113406505A - Method and device for predicting residual life of fuel cell - Google Patents
Method and device for predicting residual life of fuel cell Download PDFInfo
- Publication number
- CN113406505A CN113406505A CN202110830331.8A CN202110830331A CN113406505A CN 113406505 A CN113406505 A CN 113406505A CN 202110830331 A CN202110830331 A CN 202110830331A CN 113406505 A CN113406505 A CN 113406505A
- Authority
- CN
- China
- Prior art keywords
- fuel cell
- voltage
- acquiring
- health
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000000446 fuel Substances 0.000 title claims abstract description 296
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000036541 health Effects 0.000 claims abstract description 113
- 230000006399 behavior Effects 0.000 claims abstract description 73
- 238000003062 neural network model Methods 0.000 claims abstract description 58
- 230000004927 fusion Effects 0.000 claims abstract description 46
- 230000015654 memory Effects 0.000 claims abstract description 43
- 239000000178 monomer Substances 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 29
- 239000000498 cooling water Substances 0.000 claims abstract description 26
- 230000006403 short-term memory Effects 0.000 claims abstract description 15
- 230000007787 long-term memory Effects 0.000 claims abstract description 14
- 229910052799 carbon Inorganic materials 0.000 claims description 8
- 230000008569 process Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 7
- 239000012528 membrane Substances 0.000 description 5
- 239000007789 gas Substances 0.000 description 4
- 239000003054 catalyst Substances 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003862 health status Effects 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000002737 fuel gas Substances 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 239000005431 greenhouse gas Substances 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- BULVZWIRKLYCBC-UHFFFAOYSA-N phorate Chemical compound CCOP(=S)(OCC)SCSCC BULVZWIRKLYCBC-UHFFFAOYSA-N 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/378—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] specially adapted for the type of battery or accumulator
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/30—Hydrogen technology
- Y02E60/50—Fuel cells
Abstract
The invention discloses a method and a device for predicting the residual life of a fuel cell, wherein the fuel cell is composed of a plurality of monomers, and the method for predicting the residual life of the fuel cell comprises the following steps: acquiring parameter data of the fuel cell during the operation condition; the parameter data at least comprises voltage, current, monomer voltage, air inlet flow, air inlet pressure and cooling water inlet temperature; acquiring a fusion health index according to the parameter data; acquiring a health state value of the fuel cell according to the fusion health index; acquiring the behavior of the fuel cell during the operation condition; the behaviors at least comprise starting, stopping, load changing and fault emergency stopping; training a long-term and short-term memory neural network model according to time, a health state value and behaviors; and predicting the residual life of the fuel cell according to the long-short term memory neural network model. The technical scheme provided by the invention can realize accurate prediction of the residual service life of the fuel cell under the driving condition.
Description
Technical Field
The invention relates to the technical field of fuel cells, in particular to a method and a device for predicting the residual life of a fuel cell.
Background
The fuel cell has the advantages of less greenhouse gas emission, high energy conversion efficiency and the like, however, the vehicle-mounted fuel cell system is frequently started and stopped and has complex working conditions, and the service life of the fuel cell system is seriously influenced. Therefore, the method effectively predicts the residual service life of the fuel cell and has important significance for prolonging the service life of the fuel cell and promoting the commercial development of the fuel cell.
In the prior art, based on an endurance test of a fuel cell under a steady-state experimental condition, the remaining service life of the fuel cell is predicted by adopting a data-driven method, the influence of start-stop, load change, fault emergency stop and other behaviors on the performance attenuation of the fuel cell in an actual operation working condition is not considered, the fuel cell can generate a moving hydrogen-oxygen interface in the start-stop process, the loss of Pt catalyst particles and the corrosion of a carbon carrier can occur to a cathode after the fuel cell is started and stopped for many times, the proton membrane can be physically damaged by the load change and the fault emergency stop, the monomer resistance value distribution is uneven due to the transient operation condition change, the corrosion of the catalyst is aggravated, meanwhile, a single health index can not completely reflect the health state value of the fuel cell, the condition of one or more antipoles can occur when the voltage of the fuel cell is in an unbalanced state for a long time, and the endurance of the fuel cell is seriously influenced.
Disclosure of Invention
The embodiment of the invention provides a method and a device for predicting the residual life of a fuel cell, which are used for realizing accurate prediction of the residual life of the fuel cell under a driving condition.
In a first aspect, an embodiment of the present invention provides a method for predicting remaining life of a fuel cell, where the fuel cell is composed of a plurality of cells, and the method for predicting remaining life of a fuel cell includes:
acquiring parameter data of the fuel cell during the operation condition; the parameter data at least comprises voltage, current, cell voltage, air inlet flow, air inlet pressure and cooling water inlet temperature;
acquiring a fusion health index according to the parameter data; acquiring a health state value of the fuel cell according to the fusion health index;
acquiring the behavior of the fuel cell during the operation condition; the behaviors at least comprise starting, stopping, load changing and fault emergency stopping;
training a long-short term memory neural network model according to time, the health state values and the behaviors; and predicting the remaining life of the fuel cell according to the long-short term memory neural network model.
Optionally, extracting the parameter data of the fuel cell during the operating condition includes:
extracting a stable interval of the voltage of the fuel cell during the operating condition;
and acquiring the parameter data in the stable interval.
Optionally, obtaining the fusion health indicator according to the parameter data includes:
acquiring standard voltage V of the fuel cell at different moments in the operating condition period according to the parameter dataek(ii) a And calculating the standard voltage V of the fuel cell at the current momentekAnd the actual voltage VrealDifference Δ V ═ V betweenek-Vreal；
Obtaining the fluctuation rate C of the monomer voltage according to the monomer voltage and the number of the monomers in the fuel cellv；
Counting the cell data n of the lowest cell voltage of the fuel cell at different moments;
according to the difference value DeltaV between the standard voltage and the actual voltage of the fuel cell and the single voltage fluctuation rate CvAnd establishing the fusion health indicator HI ═ f (Δ V, n, C) from the cell data n for the lowest cell voltagev)。
Optionally, obtaining the standard voltage of the fuel cell at different times during the operating condition according to the parameter data includes:
acquiring standard voltage V of the fuel cell at different moments in the operating condition periodek＝af(I)+bf(p)+cf(W)+df(Tin) (ii) a Wherein I is current, p is air intake pressure, W is air intake flow, and T isinIs the cooling water inlet temperature.
Optionally, the cell voltage fluctuation rate C is obtained according to the cell voltage and the number of cells in the fuel cellvThe method comprises the following steps:
obtaining a cell voltage fluctuation rate in the fuel cell
Wherein, ViThe cell voltages of the respective cells of the fuel cell,
the average value of the voltage of all the monomers in the fuel cell is shown, and N is the number of the monomers of the fuel cell.
Optionally, obtaining the state of health value of the fuel cell according to the fusion health indicator includes:
acquiring a health index value HI of the fuel cell at an initial time0；
Acquiring a health index value HI of the fuel cell from an initial time to a time tt；
Acquiring the state-of-health value delta HI of the fuel cell0-HIt。
Optionally, after acquiring the behavior of the fuel cell during the operating condition, the method further includes:
marking each behavior of the fuel cell during the operation condition in a data mode; wherein the specific data in which the behavior is marked may be set according to the level of influence of the behavior on the voltage decay of the fuel cell; different data represent different impact levels.
Optionally, training the long-short term memory neural network model according to time, the health state value and the behavior, including:
inputting the time series of health state values and the behaviors at different moments into the long-short term memory neural network model;
and training the long-term and short-term memory neural network model to simulate the operation condition of the fuel cell.
Optionally, predicting the remaining life of the fuel cell according to the long-short term memory neural network model includes:
acquiring a failure threshold value of the fuel cell according to the long-short term memory neural network model;
and determining the operation time of the fuel cell when the fuel cell operates to the failure threshold value according to the long-short term memory neural network model, and determining the residual life of the fuel cell according to the operation time.
In a second aspect, an embodiment of the present invention provides a remaining life prediction apparatus for a fuel cell, including:
the parameter acquisition module is used for acquiring parameter data of the fuel cell during the operating condition; the parameter data at least comprises voltage, current, cell voltage, air inlet flow, air inlet pressure and cooling water inlet temperature;
the index fitting module is used for acquiring a fusion health index according to the parameter data; acquiring a health state value of the fuel cell according to the fusion health index;
the behavior acquisition module is used for acquiring the behavior of the fuel cell during the operation working condition period; the behaviors at least comprise starting, stopping, load changing and fault emergency stopping;
the model training module is used for training a long-term and short-term memory neural network model according to time, the health state value and the behavior;
and the life prediction module is used for predicting the residual life of the fuel cell according to the long-short term memory neural network model.
According to the technical scheme provided by the embodiment of the invention, the health state value of the fuel cell is obtained by obtaining parameter data such as voltage, current, monomer voltage, air inlet flow, air inlet pressure, cooling water inlet temperature and the like of the fuel cell during the operation working condition, obtaining the fused health index according to the parameter data, obtaining the behaviors of starting, stopping, variable load, sudden failure stop and the like of the fuel cell during the operation working condition, training the long-short term memory neural network model according to time, the health state value and the behaviors, and predicting the residual life of the fuel cell according to the long-short term memory neural network model. The method realizes accurate prediction of the residual service life of the fuel cell by fusing the health index and the behavior under the dynamic working condition of driving.
Drawings
Fig. 1 is a flowchart of a method for predicting remaining life of a fuel cell according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for predicting remaining life of a fuel cell according to another embodiment of the present invention;
fig. 3 is a flowchart of a method for predicting remaining life of a fuel cell according to another embodiment of the present invention;
FIG. 4 is a flow chart of a behavior-based long-short term memory neural network model training process according to an embodiment of the present invention;
FIG. 5 is a flow chart of another behavior-based long-short term memory neural network model training process provided by an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a device for predicting remaining life of a fuel cell according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be fully described by the detailed description with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are a part of the embodiments of the present invention, not all embodiments, and all other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without inventive efforts fall within the scope of the present invention.
Fig. 1 is a flowchart of a method for predicting remaining life of a fuel cell according to a first embodiment of the present invention, which is suitable for predicting remaining life of a fuel cell. The fuel cell is composed of a plurality of single units, and the method for predicting the residual life of the fuel cell comprises the following steps:
s101, acquiring parameter data of the fuel cell during the operation working condition; the parametric data includes at least voltage, current, cell voltage, air intake flow, air intake pressure, and cooling water inlet temperature.
The voltage is the actual voltage value of the fuel cell at the current moment, the fuel cell is equivalent to a direct current source when working, the anode of the fuel cell is the negative pole of the power supply, the cathode of the fuel cell is the positive pole of the power supply, the voltage between the anode and the cathode is the voltage in the implementation, and whether the fuel cell can be continuously used can be determined according to the voltage value.
The current refers to the current of the stack, and specifically, the fuel cell is composed of a plurality of monomers, the monomers are a sub-cell unit, and the plurality of monomers are stacked to form the fuel cell stack with output voltage meeting the actual load requirement, which is called the stack for short.
The monomer voltage refers to the voltage value of a monomer in the fuel cell at the current moment, the difference of the monomer voltage distribution of the fuel cell is objective, and when the monomer voltage of the fuel cell is in an unbalanced state for a long time, the reverse polarity of one or more monomers can occur, so that the durability of the fuel cell is seriously influenced.
The air intake flow rate refers to the mass flow rate of the gas entering the fuel cell from the anode or cathode of the fuel cell, for example, at the anode of the fuel cell, hydrogen flows into the fuel cell from the inlet, and at the cathode of the fuel cell, oxygen flows into the fuel cell from the inlet.
The air intake pressure refers to the gas pressure at which fuel gas enters the fuel cell from the anode or cathode of the fuel cell. For example, when the proton exchange membrane cell is operated, the gas pressures of the reaction gases at both sides of the proton exchange membrane should be kept relatively balanced, so that the diffusion of the fuel into the proton exchange membrane can be controlled to the minimum, and the function of protecting the proton exchange membrane can be achieved.
The cooling water inlet temperature refers to the temperature of cooling water at the inlet of the cooling device of the fuel cell, and specifically, the fuel cell generates a large amount of heat energy in the process of generating electric energy, so that the cooling device is required to absorb the heat of the fuel cell, for example, the cooling water inlet temperature of a cooling water pump may be used.
Specifically, when the fuel cell is in the operating condition, parameter data such as voltage, current, cell voltage, air intake flow, air intake pressure, cooling water inlet temperature and the like can be acquired.
It should be noted that the above parameter data are measured during the operation condition of the fuel cell, where the operation condition is the condition of the fuel cell outputting power outwards, and for example, for a vehicle using the fuel cell as a power source, when the vehicle is started or running, the operation condition is the operation condition of the fuel cell.
S102, acquiring a fusion health index according to the parameter data; and acquiring the health state value of the fuel cell according to the fusion health index.
The fusion health index refers to a non-single health index, which is a comprehensive health index established by a plurality of different data. For example, a single health index, which is comprehensively established from data such as fuel cell voltage, current, cell voltage, air intake flow rate, air intake pressure, and cooling water inlet temperature, cannot completely reflect the state of health of the fuel cell.
The state of health value refers to a value calculated by the fuel cell according to the fusion health index, and different state of health values represent different performances of the fuel cell, for example, the lower the state of health value of the fuel cell is, the worse the performance of the fuel cell is, the higher the state of health value of the fuel cell is, the better the performance of the fuel cell is.
Specifically, a fusion health index is established according to the obtained parameter data such as voltage, current, monomer voltage, air intake flow, air intake pressure, cooling water inlet temperature and the like, specific values of the parameter data of the fuel cell are obtained, and the corresponding health state value of the fuel cell can be obtained according to the fusion health index.
S103, acquiring the behavior of the fuel cell in the operating condition period; the behaviors include at least start, stop, load change and fault emergency stop.
Specifically, in actual operation of the fuel cell, there may be frequent behaviors such as starting, stopping, load changing or sudden failure stop, and the behavior of the fuel cell during the operation condition has different degrees of influence on the attenuation of the fuel cell, thereby affecting the service life of the fuel cell.
S104, training a long-term and short-term memory neural network model according to time, a health state value and behaviors; and predicting the residual life of the fuel cell according to the long-short term memory neural network model.
Wherein time refers to the time that the fuel cell is operating.
The long and short memory neural network model refers to a deep learning method for processing sequence data.
Specifically, the time, the health state value and the behavior are used as input to be trained by adopting a long and short memory neural network model to obtain a training model with accurate target, and the residual life time of the fuel cell is predicted according to the long and short memory neural network model.
According to the technical scheme provided by the embodiment of the invention, the health state value of the fuel cell is obtained by obtaining parameter data such as voltage, current, monomer voltage, air inlet flow, air inlet pressure, cooling water inlet temperature and the like of the fuel cell during the operation working condition, obtaining the fused health index according to the parameter data, obtaining the behaviors of starting, stopping, variable load, sudden failure stop and the like of the fuel cell during the operation working condition, training the long-short term memory neural network model according to time, the health state value and the behaviors, and predicting the residual life of the fuel cell according to the long-short term memory neural network model. The method realizes accurate prediction of the residual service life of the fuel cell by fusing the health index and the behavior under the dynamic working condition of driving.
Optionally, extracting parameter data of the fuel cell during the operation condition may include: extracting a stable interval of voltage of the fuel cell during the operation condition; and acquiring parameter data in the stable interval.
Wherein, the stable interval refers to the interval after the vehicle is started for a period of time. Because the voltage output by the fuel cell is unstable when the vehicle just starts, corresponding parameter data in the stable interval of the output voltage of the fuel cell is extracted, and the reliability and the validity of the data are ensured.
Fig. 2 is a flowchart of another method for predicting remaining life of a fuel cell according to an embodiment of the present invention, as shown in fig. 2, the obtaining of a fusion health indicator according to parameter data based on the above embodiment mainly includes the following steps:
s201, acquiring parameter data of the fuel cell during the operation working condition; the parametric data includes at least voltage, current, cell voltage, air intake flow, air intake pressure, and cooling water inlet temperature.
S202, acquiring standard voltage V of the fuel cell at different moments in the operating condition period according to the parameter dataek(ii) a And calculates the standard voltage V of the fuel cell at the present momentekAnd the actual voltage VrealDifference Δ V ═ V betweenek-Vreal。
The standard voltage is calculated according to the parameter data of the fuel cell in the stable interval, and the fuel cell has stable output voltage and good state.
Specifically, the standard voltage V at different moments is calculated according to the parameter data of the fuel cell in the stable intervalekObtaining actual voltage V of the fuel cell at different timesrealAnd according to the standard voltage V at the current momentekAnd the actual voltage VrealTo obtain a difference Δ V, i.e., Δ V ═ Vek-VrealThe difference Δ V may be a positive value, a negative value, or zero.
Optionally, the fuel cell is obtained according to the parameter dataThe standard voltage at different moments during the operating condition comprises: obtaining standard voltage V of fuel cell at different time points during operation conditionek＝af(I)+bf(p)+cf(W)+df(Tin) (ii) a Wherein I is current, p is air intake pressure, W is air intake flow, and T isinIs the cooling water inlet temperature.
Wherein a, b, c and d are coefficients, f (I), f (p), f (W) and f (T)in) With respect to the current I, the air intake pressure p, the air intake flow W and the cooling water inlet temperature T, respectivelyinAs a function of (c).
Specifically, according to the current I, the air intake pressure p, the air intake flow W, and the cooling water inlet temperature TinThe standard voltage V of the fuel cell at different moments can be obtainedekWhen any one variable value changes, the standard voltage V is influencedekThe value of the parameter (V) further influences the value of the delta V, finally the value of the fusion health index HI is changed, different parameter data are taken into consideration, and accurate prediction of the residual service life of the fuel cell is guaranteed.
S203, obtaining the fluctuation rate C of the single voltage according to the single voltage and the number of the single in the fuel cellv。
The number of the single cells is the specific number of the single cells in the fuel cell, and specifically, the single voltage fluctuation rate C at the current moment is calculated according to the single voltage and the number of the single cells of the fuel cellvFrom the individual voltage fluctuation rate CvIt is known that the cell voltage of the fuel cell is well balanced, and the durability of the fuel cell is better when the cell voltage is well balanced.
Optionally, the cell voltage fluctuation rate C is obtained according to the cell voltage and the number of cells in the fuel cellvThe method comprises the following steps: obtaining cell voltage fluctuation rate in fuel cell
Wherein, ViIs the cell voltage of each cell of the fuel cell,
the average value of the cell voltages of all the cells in the fuel cell, and N is the number of the cells in the fuel cell.
Wherein, the fluctuation ratio C of the cell voltagevFor evaluating cell voltage equality of fuel cells, e.g. cell voltage fluctuation ratio CvThe larger the cell voltage, the worse the cell voltage balance of the fuel cell.
Specifically, cell voltages V of the individual cells of the fuel cell are acquirediAnd the number N of the single bodies of the fuel cell, and further calculating to obtain the average value of the voltage of all the single bodies in the fuel cell
And according to the formula
Obtaining the unit voltage fluctuation rate C in the fuel cellvWherein i is more than or equal to 1 and less than or equal to N. The fluctuation ratio C of the cell voltagevAnd as the input of the fusion health index HI, the unit voltage balance is considered, so that the residual service life of the fuel cell is more accurately predicted.
And S204, counting the cell data n of the lowest cell voltage of the fuel cell at different moments.
Specifically, the cell data n of the lowest cell voltage at different times can be obtained through statistics according to the cell voltages of different cells of the fuel cell at different times. For example, the lowest cell voltage value may be set, the cell voltages of the cells at different times are obtained and compared with the lowest cell voltage value, and if the cell voltages of n cells existing at the current time are less than or equal to the lowest cell voltage value, the cell data of the lowest cell voltage is n.
S205, according to the difference value delta V between the standard voltage and the actual voltage of the fuel cell, the single voltage fluctuation ratio CvEstablishing a fusion health index HI ═ f (delta V, n, C) with the cell data n of the lowest cell voltagev)。
Wherein the fusion health index is represented by HI, f (delta V, n, C)v) Mean difference value DeltaV and monomer voltage fluctuation ratio CvAnd lowest cell voltageAnd the monomer data n form a functional relation.
Specifically, according to the difference value Δ V and the cell voltage fluctuation ratio CvEstablishing a fusion health index HI with the cell data n of the lowest cell voltage, in other words, the fusion health index HI is related to the difference value DeltaV and the cell voltage fluctuation rate CvAnd the cell data n of the lowest cell voltage, any change in either variable will affect the value of the fusion health indicator HI.
The above steps S202 to S205 are a specific process of "obtaining the fusion health index according to the parameter data", so that a plurality of health indexes are comprehensively considered to realize accurate evaluation of the current health state of the fuel cell.
And S206, acquiring the health state value of the fuel cell according to the fusion health index.
S207, acquiring the behavior of the fuel cell in the operating condition period; the behaviors include at least start, stop, load change and fault emergency stop.
S208, training a long-term and short-term memory neural network model according to time, health state values and behaviors; and predicting the residual life of the fuel cell according to the long-short term memory neural network model.
The embodiment specifically details the process of obtaining the fusion health index according to the parameter data, and the fluctuation rate C of the cell voltage is obtained by the difference Δ V between the standard voltage and the actual voltage of the fuel cellvAnd establishing the fusion health index, fusion current I, air inlet pressure p, air inlet flow W and cooling water inlet temperature T according to the monomer data n of the lowest monomer voltageinAnd the data of various parameters are beneficial to completely reflecting the health state of the fuel cell.
Fig. 3 is a flowchart of a method for predicting remaining life of a fuel cell according to an embodiment of the present invention, as shown in fig. 3, the obtaining a state of health value of the fuel cell according to a fusion health indicator mainly includes the following steps:
s301, acquiring parameter data of the fuel cell during the operation working condition; the parameter data at least comprises voltage, current, monomer voltage, air inlet flow, air inlet pressure and cooling water inlet temperature;
s302, acquiring standard voltage V of the fuel cell at different moments in the operating condition period according to the parameter dataek(ii) a And calculates the standard voltage V of the fuel cell at the present momentekAnd the actual voltage VrealDifference Δ V ═ V betweenek-Vreal。
S303, obtaining the fluctuation rate C of the single voltage according to the single voltage and the number of the single in the fuel cellv。
And S304, counting the cell data n of the lowest cell voltage of the fuel cell at different moments.
S305, according to the difference value delta V between the standard voltage and the actual voltage of the fuel cell, the single voltage fluctuation ratio CvEstablishing a fusion health index HI ═ f (delta V, n, C) with the cell data n of the lowest cell voltagev)。
S306, acquiring a health index value HI of the fuel cell at the initial moment0。
The initial time refers to a health index value at the time of factory setting of the fuel cell.
Specifically, the initial time t of the fuel cell is acquired0Time voltage, current, monomer voltage, air inlet flow, air inlet pressure, cooling water inlet temperature and other parameter data are calculated according to the steps in sequence to obtain the monomer voltage fluctuation rate C at the current momentvThe difference Δ V and the cell data n for the lowest cell voltage, and further according to the fusion health index HI ═ f (Δ V, n, C)v) Obtaining a health index value HI of the fuel cell at an initial time0。
S307, acquiring a health index value HI of the fuel cell from the initial time to the t timet。
time t is the fuel cell from the initial time t0At the beginning of the operation, the fuel cell is put into use for a certain period of time, for example, the last half year or one year of the vehicle, specifically, the parameter data of voltage, current, cell voltage, air intake flow, air intake pressure and cooling water inlet temperature at the time of t of the fuel cell are obtained, and the cell voltage fluctuation rate C at the current time is obtained by calculating according to the stepsvThe difference Δ V and the lowestThe cell data n of the cell voltage, and in turn HI ═ f (Δ V, n, C) according to the fusion health indexv) Obtaining a health index value HI at time t of the fuel cellt。
S308, acquiring the state of health value delta HI of the fuel cell0-HIt。
Specifically, the health index value HI at the initial time of the fuel cell obtained in step S306 and step S3070And a health index value HI at time ttThe state of health value of the fuel cell, i.e., Δ HI — HI, can be calculated0-HIt。
The above steps S306 to S308 are specific procedures of "obtaining the health state value of the fuel cell according to the fusion health indicator". Therefore, the state of health value of the fuel cell in any time period can be obtained according to the specific implementation process of the state of health value of the fuel cell, the current state of health of the fuel cell can be accurately predicted, and the real-time monitoring of the fuel cell under the driving condition is realized.
S309, acquiring the behavior of the fuel cell in the operation working condition period; the behaviors at least comprise starting, stopping, load changing and fault emergency stopping;
with continued reference to fig. 3, optionally, after obtaining the behavior of the fuel cell during the operating condition, the method further comprises the steps of:
s310, marking each behavior of the fuel cell in the operation working condition in a data mode; wherein the specific data marked by the behavior can be set according to the influence level of the behavior on the voltage attenuation of the fuel cell; different data represent different impact levels.
The digitization refers to a quantized mode, for example, the fuel cell is represented by binary number 0000 when it is not behaving during the operation condition, the start is represented by binary number 0001, the stop is represented by binary number 0010, the load change is represented by binary number 0011, and the fault emergency stop is represented by binary number 0100, and the specific digitized marking mode is not limited in the embodiment of the invention.
The voltage drop of the fuel cell means a decrease in the output voltage of the fuel cell, and may be, for example, a decrease in the voltage at the same power, an imbalance of cells, an increase in the lowest cell voltage number, or the like, which affects the voltage drop of the fuel cell.
Specifically, each behavior of the fuel cell during the operation condition is marked in a digitalized manner, and the marked specific data can be set according to the level of the influence of the behavior on the voltage attenuation of the fuel cell, for example, the larger the value of the marked specific data is, the larger the influence of the behavior on the voltage attenuation of the fuel cell is represented. Therefore, the influence of behaviors on the voltage attenuation of the fuel cell can be judged according to the size of the marked specific data, so that the fuel cell system can conveniently analyze and process the marked specific data according to each behavior, a training model which is more in line with the actual situation is further established, and the target accuracy of the training model is improved.
S311, training a long-term and short-term memory neural network model according to time, a health state value and behaviors; and predicting the residual life of the fuel cell according to the long-short term memory neural network model.
Optionally, predicting the remaining life of the fuel cell according to the long-term short-term memory neural network model includes: acquiring a failure threshold value of the fuel cell according to the long-short term memory neural network model; and determining the operation time of the fuel cell when the fuel cell operates to the failure threshold according to the long-short term memory neural network model, and determining the residual life of the fuel cell according to the operation time.
The failure threshold refers to a lowest output voltage value of the fuel cell during the operation condition, and may be, for example, a factory set value of the fuel cell, and specifically may be 20% of a rated output voltage of the fuel cell. When the voltage output by the fuel cell during operating conditions reaches a failure threshold, the fuel cell will not continue to be used.
The operation duration refers to the time elapsed from the time when the fuel cell starts to operate after leaving the factory to the time when the voltage output by the fuel cell reaches the failure threshold.
Specifically, the failure threshold of the fuel cell can be obtained according to the long-short term memory neural network model, the operation time of the fuel cell when the fuel cell operates to the failure threshold can be calculated according to the long-short term memory neural network model, and the remaining service life of the fuel cell can be further determined. And a long-term and short-term memory neural network model is adopted to perform time sequence data processing, so that a failure threshold value and operation duration are accurately obtained, and the accurate prediction of the residual service life of the fuel cell is achieved.
Fig. 4 is a flowchart of training a long-short term memory neural network model based on behaviors according to an embodiment of the present invention, and fig. 5 is a flowchart of training a long-short term memory neural network model based on behaviors according to an embodiment of the present invention, which is combined with the flowchart shown in fig. 4 and fig. 5, and which trains the long-short term memory neural network model according to time, health status values and behaviors, including the following steps:
s401, inputting the time sequence of the health state values and behaviors at different moments into a long-term and short-term memory neural network model;
specifically, the health state value of the fuel cell and each behavior of the fuel cell at different moments during the operation working condition are obtained through calculation by obtaining the fusion health index HI, the health state value and each behavior are used as the input of the long-short term memory neural network model, and the long-short term memory neural network model is trained.
S402, training the long-term and short-term memory neural network model to enable the model to simulate the operation condition of the fuel cell.
Specifically, as shown in fig. 5, network parameters and weights in the long-term and short-term memory neural network model are set, and the weights are updated based on a gradient descent method until a training model with an accurate target is obtained, at which time, training of the network model is completed.
The trained long-term and short-term memory neural network model can form a model of the fuel cell, and the operation condition and the working state of the model approach the actual condition of the fuel cell. The failure threshold value of the fuel cell can be obtained according to the long-short term memory neural network model, the operation time of the fuel cell when the fuel cell operates to the failure threshold value can be calculated according to the long-short term memory neural network model, and the remaining service life of the fuel cell can be accurately determined.
Based on the same idea, the present invention further provides a device for predicting remaining life of a fuel cell, as shown in fig. 6, fig. 6 is a schematic structural diagram of a device for predicting remaining life of a fuel cell according to an embodiment of the present invention, and the device includes: the parameter acquisition module 601 is used for acquiring parameter data of the fuel cell during the operation condition; the parameter data at least comprises voltage, current, monomer voltage, air inlet flow, air inlet pressure and cooling water inlet temperature; an index fitting module 602, configured to obtain a fusion health index according to the parameter data; acquiring a health state value of the fuel cell according to the fusion health index; a behavior acquisition module 603 for acquiring the behavior of the fuel cell during the operating condition; the behaviors at least comprise starting, stopping, load changing and fault emergency stopping; a model training module 604 for training the long-short term memory neural network model according to time, health status values and behaviors; and a life prediction module 605 for predicting the remaining life of the fuel cell according to the long-short term memory neural network model.
In the embodiment of the invention, parameter data of the fuel cell during the operation working condition is obtained by setting a parameter obtaining module; the index fitting module acquires a fusion health index according to the parameter data and acquires a health state value of the fuel cell according to the fusion health index; the behavior acquisition module acquires the behavior of the fuel cell during the operation condition; the model training module trains the long-term and short-term memory neural network model according to time, the health state value and the behavior; the life prediction module predicts a remaining life of the fuel cell based on the long-short term memory neural network model. The method realizes accurate prediction of the remaining service life of the fuel cell based on fusion of health indexes and behaviors under the dynamic working condition of driving.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method of predicting remaining life of a fuel cell, the fuel cell being composed of a plurality of cells, the method comprising:
acquiring parameter data of the fuel cell during the operation condition; the parameter data at least comprises voltage, current, cell voltage, air inlet flow, air inlet pressure and cooling water inlet temperature;
acquiring a fusion health index according to the parameter data; acquiring a health state value of the fuel cell according to the fusion health index;
acquiring the behavior of the fuel cell during the operation condition; the behaviors at least comprise starting, stopping, load changing and fault emergency stopping;
training a long-short term memory neural network model according to time, the health state values and the behaviors; and predicting the remaining life of the fuel cell according to the long-short term memory neural network model.
2. The method of predicting remaining life of a fuel cell according to claim 1, wherein extracting the parameter data of the fuel cell during the operating condition includes:
extracting a stable interval of the voltage of the fuel cell during the operating condition;
and acquiring the parameter data in the stable interval.
3. The method of predicting remaining life of a fuel cell according to claim 1, wherein obtaining the fusion health indicator from the parameter data includes:
acquiring standard voltage V of the fuel cell at different moments in the operating condition period according to the parameter dataek(ii) a And calculating the standard voltage V of the fuel cell at the current momentekAnd the actual voltage VrealDifference Δ V ═ V betweenek-Vreal；
Obtaining the fluctuation rate C of the monomer voltage according to the monomer voltage and the number of the monomers in the fuel cellv；
Counting the cell data n of the lowest cell voltage of the fuel cell at different moments;
according to the difference value DeltaV between the standard voltage and the actual voltage of the fuel cell and the single voltage fluctuation rate CvAnd establishing the fusion health indicator HI ═ f (Δ V, n, C) from the cell data n for the lowest cell voltagev)。
4. The method of predicting the remaining life of the fuel cell according to claim 3, wherein obtaining the standard voltage of the fuel cell at different times during the operating condition based on the parameter data includes:
acquiring standard voltage V of the fuel cell at different moments in the operating condition periodek＝af(I)+bf(p)+cf(W)+df(Tin) (ii) a Wherein I is current, p is air intake pressure, W is air intake flow, and T isinIs the cooling water inlet temperature.
5. The method according to claim 3, wherein the cell voltage fluctuation rate C is obtained from the cell voltage and the number of cells in the fuel cellvThe method comprises the following steps:
obtaining a cell voltage fluctuation rate in the fuel cell
Wherein, ViThe cell voltages of the respective cells of the fuel cell,
the average value of the voltage of all the monomers in the fuel cell is shown, and N is the number of the monomers of the fuel cell.
6. The method of predicting remaining life of a fuel cell according to claim 1, wherein obtaining the state of health value of the fuel cell based on the fusion health indicator includes:
acquiring a health index value HI of the fuel cell at an initial time0；
Acquiring a health index value HI of the fuel cell from an initial time to a time tt；
Acquiring the state-of-health value delta HI of the fuel cell0-HIt。
7. The method of predicting the remaining life of the fuel cell according to claim 1, further comprising, after acquiring the behavior of the fuel cell during the operating condition:
marking each behavior of the fuel cell during the operation condition in a data mode; wherein the specific data in which the behavior is marked may be set according to the level of influence of the behavior on the voltage decay of the fuel cell; different data represent different impact levels.
8. The method of predicting remaining life of a fuel cell according to claim 1, wherein training a long-short term memory neural network model according to time, the state of health value, and the behavior comprises:
inputting the time series of health state values and the behaviors at different moments into the long-short term memory neural network model;
and training the long-term and short-term memory neural network model to simulate the operation condition of the fuel cell.
9. The method of predicting the remaining life of the fuel cell according to claim 1, wherein predicting the remaining life of the fuel cell based on the long-short term memory neural network model comprises:
acquiring a failure threshold value of the fuel cell according to the long-short term memory neural network model;
and determining the operation time of the fuel cell when the fuel cell operates to the failure threshold value according to the long-short term memory neural network model, and determining the residual life of the fuel cell according to the operation time.
10. A remaining life prediction apparatus for a fuel cell, comprising:
the parameter acquisition module is used for acquiring parameter data of the fuel cell during the operation working condition; the parameter data at least comprises voltage, current, cell voltage, air inlet flow, air inlet pressure and cooling water inlet temperature;
the index fitting module is used for acquiring a fusion health index according to the parameter data; acquiring a health state value of the fuel cell according to the fusion health index;
the behavior acquisition module is used for acquiring the behavior of the fuel cell during the operation working condition period; the behaviors at least comprise starting, stopping, load changing and fault emergency stopping;
the model training module is used for training a long-term and short-term memory neural network model according to time, the health state value and the behavior;
and the life prediction module is used for predicting the residual life of the fuel cell according to the long-short term memory neural network model.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110830331.8A CN113406505A (en) | 2021-07-22 | 2021-07-22 | Method and device for predicting residual life of fuel cell |
| PCT/CN2022/104187 WO2023000985A1 (en) | 2021-07-22 | 2022-07-06 | Method and apparatus for predicting remaining life of fuel cell |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110830331.8A CN113406505A (en) | 2021-07-22 | 2021-07-22 | Method and device for predicting residual life of fuel cell |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113406505A true CN113406505A (en) | 2021-09-17 |
Family
ID=77687316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110830331.8A Pending CN113406505A (en) | 2021-07-22 | 2021-07-22 | Method and device for predicting residual life of fuel cell |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113406505A (en) |
| WO (1) | WO2023000985A1 (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114142069A (en) * | 2021-11-26 | 2022-03-04 | 广东电网有限责任公司广州供电局 | Online monitoring device and method based on fuel cell health state evaluation |
| CN114759227A (en) * | 2022-05-07 | 2022-07-15 | 中国第一汽车股份有限公司 | Method and device for determining degradation of fuel cell performance |
| CN115453369A (en) * | 2022-09-20 | 2022-12-09 | 中国汽车工程研究院股份有限公司 | Method for fuel cell consistency prediction and fault diagnosis |
| WO2023000985A1 (en) * | 2021-07-22 | 2023-01-26 | 中国第一汽车股份有限公司 | Method and apparatus for predicting remaining life of fuel cell |
| WO2023045914A1 (en) * | 2021-09-23 | 2023-03-30 | 中国第一汽车股份有限公司 | Fuel cell control method and control apparatus, device, and storage medium |
| CN116068413A (en) * | 2023-03-22 | 2023-05-05 | 长安新能源南京研究院有限公司 | Battery voltage prediction method, device, equipment and storage medium |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116502544B (en) * | 2023-06-26 | 2023-09-12 | 武汉新威奇科技有限公司 | Electric screw press life prediction method and system based on data fusion |
| CN116995276B (en) * | 2023-09-27 | 2023-12-29 | 爱德曼氢能源装备有限公司 | Cooling method and system for fuel cell power generation system |
| CN117350174B (en) * | 2023-12-04 | 2024-04-02 | 国网天津市电力公司营销服务中心 | Method, system, electronic equipment and medium for predicting residual life of intelligent ammeter |
| CN117491793A (en) * | 2023-12-29 | 2024-02-02 | 国网浙江省电力有限公司电力科学研究院 | Hydrogen electric coupling system comprehensive performance test method, device and medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108417868A (en) * | 2018-03-05 | 2018-08-17 | 中国第汽车股份有限公司 | A kind of vehicle fuel battery service life accelerated test and analysis method |
| EP3503274A1 (en) * | 2017-12-11 | 2019-06-26 | Commissariat à l'Energie Atomique et aux Energies Alternatives | Method for estimating the state of health of a fuel cell from measurements in actual use |
| CN109991542A (en) * | 2019-03-27 | 2019-07-09 | 东北大学 | Lithium ion battery residual life prediction technique based on WDE optimization LSTM network |
| CN110059377A (en) * | 2019-04-02 | 2019-07-26 | 西南交通大学 | A kind of fuel battery service life prediction technique based on depth convolutional neural networks |
| CN110929451A (en) * | 2019-10-24 | 2020-03-27 | 同济大学 | Fuel cell single voltage consistency prediction method |
| CN111103544A (en) * | 2019-12-26 | 2020-05-05 | 江苏大学 | Lithium ion battery remaining service life prediction method based on long-time and short-time memory LSTM and particle filter PF |
| CN111146478A (en) * | 2019-12-22 | 2020-05-12 | 同济大学 | Method for predicting residual service life of proton exchange membrane fuel cell stack |
| CN111339712A (en) * | 2020-02-24 | 2020-06-26 | 电子科技大学 | Method for predicting residual life of proton exchange membrane fuel cell |
| CN111443294A (en) * | 2020-04-10 | 2020-07-24 | 华东理工大学 | Method and device for indirectly predicting remaining life of lithium ion battery |
| KR102216850B1 (en) * | 2019-12-05 | 2021-02-18 | (주)프라즈마 사이언스 | Fuel cell management system based on artificial intelligence |
| CN112540317A (en) * | 2020-12-16 | 2021-03-23 | 武汉理工大学 | Battery health state estimation and residual life prediction method based on real vehicle data |
| CN112926273A (en) * | 2021-04-13 | 2021-06-08 | 中国人民解放军火箭军工程大学 | Method for predicting residual life of multivariate degradation equipment |
| CN112986827A (en) * | 2021-04-12 | 2021-06-18 | 山东凯格瑞森能源科技有限公司 | Fuel cell residual life prediction method based on deep learning |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103698709B (en) * | 2013-12-11 | 2016-01-13 | 清华大学 | A kind of fuel cell method for predicting residual useful life |
| CN110190306B (en) * | 2019-06-04 | 2022-08-05 | 昆山知氢信息科技有限公司 | Online fault diagnosis method for fuel cell system |
| CN111880100A (en) * | 2020-08-07 | 2020-11-03 | 同济大学 | Fuel cell residual life prediction method based on adaptive extended Kalman filtering |
| CN112784216B (en) * | 2021-01-26 | 2022-12-16 | 武汉船用电力推进装置研究所（中国船舶重工集团公司第七一二研究所） | Health degree evaluation method and system of proton exchange membrane fuel cell system |
| CN113406505A (en) * | 2021-07-22 | 2021-09-17 | 中国第一汽车股份有限公司 | Method and device for predicting residual life of fuel cell |
-
2021
- 2021-07-22 CN CN202110830331.8A patent/CN113406505A/en active Pending
-
2022
- 2022-07-06 WO PCT/CN2022/104187 patent/WO2023000985A1/en unknown
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3503274A1 (en) * | 2017-12-11 | 2019-06-26 | Commissariat à l'Energie Atomique et aux Energies Alternatives | Method for estimating the state of health of a fuel cell from measurements in actual use |
| CN108417868A (en) * | 2018-03-05 | 2018-08-17 | 中国第汽车股份有限公司 | A kind of vehicle fuel battery service life accelerated test and analysis method |
| WO2020191800A1 (en) * | 2019-03-27 | 2020-10-01 | 东北大学 | Method for predicting remaining service life of lithium-ion battery employing wde-optimized lstm network |
| CN109991542A (en) * | 2019-03-27 | 2019-07-09 | 东北大学 | Lithium ion battery residual life prediction technique based on WDE optimization LSTM network |
| CN110059377A (en) * | 2019-04-02 | 2019-07-26 | 西南交通大学 | A kind of fuel battery service life prediction technique based on depth convolutional neural networks |
| CN110929451A (en) * | 2019-10-24 | 2020-03-27 | 同济大学 | Fuel cell single voltage consistency prediction method |
| KR102216850B1 (en) * | 2019-12-05 | 2021-02-18 | (주)프라즈마 사이언스 | Fuel cell management system based on artificial intelligence |
| CN111146478A (en) * | 2019-12-22 | 2020-05-12 | 同济大学 | Method for predicting residual service life of proton exchange membrane fuel cell stack |
| CN111103544A (en) * | 2019-12-26 | 2020-05-05 | 江苏大学 | Lithium ion battery remaining service life prediction method based on long-time and short-time memory LSTM and particle filter PF |
| CN111339712A (en) * | 2020-02-24 | 2020-06-26 | 电子科技大学 | Method for predicting residual life of proton exchange membrane fuel cell |
| CN111443294A (en) * | 2020-04-10 | 2020-07-24 | 华东理工大学 | Method and device for indirectly predicting remaining life of lithium ion battery |
| CN112540317A (en) * | 2020-12-16 | 2021-03-23 | 武汉理工大学 | Battery health state estimation and residual life prediction method based on real vehicle data |
| CN112986827A (en) * | 2021-04-12 | 2021-06-18 | 山东凯格瑞森能源科技有限公司 | Fuel cell residual life prediction method based on deep learning |
| CN112926273A (en) * | 2021-04-13 | 2021-06-08 | 中国人民解放军火箭军工程大学 | Method for predicting residual life of multivariate degradation equipment |
Non-Patent Citations (4)
| Title |
|---|
| KUI CHEN, ET AL: "Proton Exchange Membrane Fuel Cell Degradation and Remaining Useful Life Prediction based on Artificial Neural Network", 《7TH INTERNATIONAL CONFERENCE ON RENEWABLE ENERGY RESEARCH AND APPLICATIONS》, 31 October 2018 (2018-10-31), pages 407 - 411, XP033469813, DOI: 10.1109/ICRERA.2018.8567023 * |
| WANG YP, ET AL: "A Data-Driven Approach to Lifespan Prediction for Vehicle Fuel Cell Systems", 《IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION》, vol. 9, no. 4, 31 December 2023 (2023-12-31), pages 5049 - 5060, XP011956480, DOI: 10.1109/TTE.2023.3237219 * |
| 李奇;刘嘉蔚;陈维荣;: "质子交换膜燃料电池剩余使用寿命预测方法综述及展望", 中国电机工程学报, no. 08, 27 February 2019 (2019-02-27), pages 2365 - 2375 * |
| 马秋玉等: "燃料电池行业标准现状综述", 《汽车文摘》, no. 1, 31 January 2020 (2020-01-31), pages 14 - 17 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023000985A1 (en) * | 2021-07-22 | 2023-01-26 | 中国第一汽车股份有限公司 | Method and apparatus for predicting remaining life of fuel cell |
| WO2023045914A1 (en) * | 2021-09-23 | 2023-03-30 | 中国第一汽车股份有限公司 | Fuel cell control method and control apparatus, device, and storage medium |
| CN114142069A (en) * | 2021-11-26 | 2022-03-04 | 广东电网有限责任公司广州供电局 | Online monitoring device and method based on fuel cell health state evaluation |
| CN114759227A (en) * | 2022-05-07 | 2022-07-15 | 中国第一汽车股份有限公司 | Method and device for determining degradation of fuel cell performance |
| CN115453369A (en) * | 2022-09-20 | 2022-12-09 | 中国汽车工程研究院股份有限公司 | Method for fuel cell consistency prediction and fault diagnosis |
| CN116068413A (en) * | 2023-03-22 | 2023-05-05 | 长安新能源南京研究院有限公司 | Battery voltage prediction method, device, equipment and storage medium |
| CN116068413B (en) * | 2023-03-22 | 2023-08-18 | 长安新能源南京研究院有限公司 | Battery voltage prediction method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023000985A1 (en) | 2023-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113406505A (en) | Method and device for predicting residual life of fuel cell | |
| CN111146478B (en) | Method for predicting residual service life of proton exchange membrane fuel cell stack | |
| CN107180983B (en) | Fault diagnosis method and system for solid oxide fuel cell stack | |
| JP4895989B2 (en) | Online detection of stack crossover rate for an adaptive hydrogen abstraction framework | |
| CN108417868B (en) | Accelerated test and analysis method for service life of vehicle fuel cell | |
| CN110988723B (en) | LSTM-based battery internal resistance prediction and fault early warning method | |
| CN112289385B (en) | Electrochemical impedance spectrum prediction method for high-power proton exchange membrane fuel cell stack | |
| CN112327167A (en) | Battery capacity diving risk assessment method and system | |
| WO2013083872A1 (en) | Method and arrangement for indicating solid oxide cell operating conditions | |
| CN110749825B (en) | Method and device for establishing acceleration working condition of fuel cell durability test | |
| KR20120096614A (en) | Method for detecting fail of hydrogen supply system for fuel cell | |
| CN117117258B (en) | Fault monitoring method and device for hydrogen fuel cell system | |
| CN117558942A (en) | Method and system for detecting fuel cell stack offline based on intelligent network connection | |
| WO2013083873A1 (en) | Method and arrangement for diagnosis of solid oxide cells operating conditions | |
| CN105487017B (en) | A kind of substation UPS valve controlled sealed lead-acid accumulator state estimation and Forecasting Methodology | |
| Jacome et al. | A review of model-based prognostic for proton exchange membrane fuel cell under automotive load cycling | |
| CN116381541A (en) | Health assessment method and system for energy storage lithium battery system | |
| CN105759217A (en) | Lead-acid battery pack online fault diagnosis method based on measurable data | |
| CN114924202A (en) | Method and device for detecting service life of fuel cell | |
| Cao et al. | A neural network based method for real-time measurement of capacity and SOC of vanadium redox flow battery | |
| CN112713291A (en) | Fuel cell system and control method thereof | |
| Tang et al. | Adaptive Gaussian process regression based remaining useful life prediction of PEMFC incorporating an improved health indicator | |
| CN117217031B (en) | Intelligent production method and system for fuel cell stack | |
| Hähnel et al. | Online identification of an electric PEMFC model for power control by NMPC | |
| JP2007305404A (en) | Fuel cell stack deterioration acceleration test method, and device required therewith |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |